MCQ Book PDF

ORACLE

Color profile: Generic CMYK printer Series

profile / Oracle8i Certified Professional DBA Practice Exams / Couchman / 3341-4 / Chapter 1
Composite Default screen
Blind Folio 1:3
CHAPTER
1
OCP Exam 1:
Introduction to SQL
and PL/SQL

P:\010Comp\Oracle8\341-4\ch01.vp
Thursday, February 08, 2001 2:11:42 PM
ORACLE
Blind Folio 1:4
4 Oracle8i Certified Professional DBA Practice Exams
CP Exam 1 in the Oracle DBA track covers concepts and practices
O around the use of SQL and PL/SQL commands. To pass this exam,
you need to demonstrate an understanding of the basic SQL
constructs available in Oracle, including built-in functions. You
should also understand the basic concepts behind an Oracle
relational database management system. In more recent editions of OCP Exam 1,
the focus has included understanding of use of the PL/SQL programming language.
In addition, new features in PL/SQL introduced in Oracle8i are tested, so you should
also be sure you understand these new features.
OCP Exam 1 Topic Areas

The following topic areas are covered in OCP Exam 1. Note that these concepts
are taken directly from the Oracle OCP Candidate Guide for OCP Exam 1 and are
current as of the publication of this book. The topics and subtopics are as follows:
1. Overview of Relational Databases, SQL, and PL/SQL

1.1. Discuss the theoretical and physical aspects of a relational database
1.2. Describe the Oracle implementation of the RDBMS and ORDBMS
1.3. Describe the use and benefits of PL/SQL
2. Writing Basic SQL Statements
2.1. List the capabilities of SQL select statements
2.2. Execute a basic select statement
2.3. Differentiate between SQL statements and SQL*Plus commands
3. Restricting and Sorting Data
3.1. Limit the rows retrieved by a query
3.2. Sort the rows retrieved by a query
4. Single-Row Functions
4.1. Describe various types of functions available in SQL
4.2. Use character, number, and date functions in select statements
4.3. Describe the use of conversion functions
ORACLE
Blind Folio 1:5
Chapter 1: OCP Exam 1: Introduction to SQL and PL/SQL 5
5. Displaying Data from Multiple Tables

5.1. Write select statements to access data from more than one
table using equality and nonequality joins
5.2. View data that generally does not meet a join condition by
using outer joins
5.3. Join a table to itself
6. Aggregating Data Using Group Functions
6.1. Identify the available group functions
6.2. Describe the use of group functions
6.3. Group data using the group by clause
6.4. Include or exclude grouped rows using the having clause
7. Subqueries
7.1. Describe the types of problems that subqueries can solve
7.2. Define subqueries
7.3. List the type of subqueries
7.4. Write single-row and multiple-row subqueries
8. Multiple-Column Subqueries
8.1. Write multiple-column subqueries
8.2. Describe and maintain the behavior of subqueries when
NULL values are retrieved
8.3. Write subqueries in a from clause
9. Producing Readable Output from SQL*Plus
9.1. Produce queries that require an input variable
9.2. Customize the SQL*Plus environment
9.3. Produce more readable output
9.4. Create and execute script files
9.5. Save customizations
ORACLE
Blind Folio 1:6
10. Manipulating Data

10.1. Describe each DML statement
10.2. insert rows into a table
10.3. update rows in a table
10.4. delete rows from a table
10.5. Control transactions
11. Creating and Managing Tables
11.1. Describe the main database objects
11.2. Create tables
11.3. Describe the datatypes that can be used when specifying
column definition
11.4. Alter table definitions
11.5. Drop, rename, and truncate tables
12. Including Constraints
12.1. Describe constraints
12.2. Create and maintain constraints
13. Creating Views
13.1. Describe a view
13.2. Create a view
13.3. Retrieve data through a view
13.4. insert, update, or delete data through a view
13.5. Drop a view
14. Oracle Data Dictionary
14.1. Describe the data dictionary views a user may access
14.2. Query data from the data dictionary
15. Other Database Objects
15.1. Describe database objects and their uses
15.2. Create, maintain, and use sequences
ORACLE
Blind Folio 1:7
15.3. Create and maintain indexes

15.4. Create public and private synonyms
16. Controlling User Access
16.1. Create users
16.2. Create roles to ease setup and maintenance of the security model
16.3. Use the grant and revoke commands to grant and revoke
object privileges
17. Declaring Variables
17.1. List the benefits of PL/SQL
17.2. Describe the basic PL/SQL block and its sections
17.3. Describe the significance of variables in PL/SQL
17.4. Declare PL/SQL variables
17.5. Execute a PL/SQL block
18. Write Executable Statements
18.1. Describe the significance of the executable section
18.2. Write statements in the executable section
18.3. Describe the rules of nested blocks
18.4. Execute and test a PL/SQL block
18.5. Use coding conventions
19. Interacting with the Oracle Server
19.1. Write a successful select statement in PL/SQL
19.2. Declare the datatype and size of the PL/SQL variable dynamically
19.3. Write DML statements in PL/SQL
19.4. Control transactions in PL/SQL
19.5. Determine the outcome of SQL statements
20. Writing Control Structures
20.1. Identify the uses and types of control structures
20.2. Construct an if statement
ORACLE
Blind Folio 1:8
20.3. Construct and identify different loop statements

20.4. Use logic tables
20.5. Control block flow using nested loops and labels
21. Working with Composite Datatypes
21.1. Create user-defined PL/SQL records
21.2. Create a record with the %rowtype attribute
21.3. Create a PL/SQL table
21.4. Create a PL/SQL table of records
21.5. Describe the difference between records, tables, and tables of records
22. Writing Explicit Cursors
22.1. Distinguish between an implicit and an explicit cursor
22.2. Use a PL/SQL record variable
22.3. Write a cursor for loop
23. Advanced Explicit Cursor Concepts
23.1. Write a cursor that uses parameters
23.2. Determine when a for update clause in a cursor is required
23.3. Determine when to use the where current of clause
23.4. Write a cursor that uses a subquery
24. Handling Exceptions
24.1. Define PL/SQL exceptions
24.2. Recognize unhandled exceptions
24.3. List and use different types of PL/SQL exception handlers
24.4. Trap unanticipated errors
24.5. Describe the effect of exception propagation in nested blocks
24.6. Customize PL/SQL exception messages
ORACLE
Blind Folio 1:9
Practice Exam 1
1. You are formulating a SQL statement to retrieve data from Oracle.
Which of the following SQL statements are invalid?
A. select NAME,Jersey_No where Jersey_No = 6;
B. select NAME, Jersey_No from PLAYERS;
C. select * from PLAYERS where Jersey_No = 6;
D. select Jersey_No from PLAYERS;
2. You can use the PL/SQL block example to answer the following question:
DECLARE
CURSOR My_Employees IS
SELECT name, title FROM employee;
My_Name VARCHAR2(30);
My_Title VARCHAR2(30);
BEGIN
OPEN My_Employees;
LOOP
FETCH My_Employees INTO My_Name, My_Title;
EXIT WHEN My_Employees%NOTFOUND;
INSERT INTO MY_EMPS (MY_EMPNAME, MY_EMPTITLE)
VALUES (My_Name, My_Title);
END LOOP;
CLOSE My_Employees;
END;
If you were rewriting this block of PL/SQL, which of the following types
of loops would you use if you wanted to reduce the amount of code by
utilizing features of the loop that handle mundane aspects of processing
automatically?
A. loop ... exit when
B. while ... loop
C. loop ... loop … end
D. cursor for loop
ORACLE
Blind Folio 1:10
3. You are coding a complex PL/SQL block where several procedures call
other procedures. You have one outermost procedure that calls all other
procedures. If you wanted to prevent the user of the outermost procedure
from having the procedure fail due to an unanticipated problem, you
would include which of the following exceptions?
A. no_data_found
B. others
C. zero_divide
D. too_many_rows
4. You are attempting to develop a more robust PL/SQL application. Which
of the following keywords allow you to associate a user-defined error
message with an exception condition?
A. pragma
B. others
C. raise_application_error
D. exception
5. You are processing some data changes in your SQL*Plus session as part
of one transaction. Which of the following choices does not typically
indicate the end of a transaction?
A. Issuing an update statement
B. Issuing a commit statement
C. Issuing a rollback statement
D. Ending your session
6. You have just removed 1,700 rows from a table that were no longer
needed. In order to save the changes you’ve made to the database,
which of the following statements are used?
A. savepoint
B. commit
C. rollback
D. set transaction
ORACLE
Blind Folio 1:11
7. To identify the columns that are indexed exclusively as the result of their
inclusion in a constraint, which of the following dictionary views would
be appropriate?
A. USER_INDEXES
B. USER_TAB_COLUMNS
C. USER_COLUMNS
D. USER_CONS_COLUMNS
8. You are creating some tables in your database as part of the logical data
model. Which of the following constraints have an index associated with
them that is generated automatically by Oracle?
A. UNIQUE
B. FOREIGN KEY
C. CHECK
D. NOT NULL
9. You have a table with three associated indexes, two triggers, two
references to that table from other tables, and a view. You issue the
drop table cascade constraints statement. Which of the following
objects will still remain after the statement is issued?
A. The triggers
B. The indexes
C. The foreign keys in the other tables
D. The view
10. You are using SQL operations in Oracle. All of the following DATE
functions return a DATE datatype, except one. Which one is it?
A. NEW_TIME
B. LAST_DAY
C. ADD_MONTHS
D. MONTHS_BETWEEN
ORACLE
Blind Folio 1:12
11. You issue a select statement on the BANK_ACCT table containing the
order by clause. Which of the following uses of the order by clause
would produce an error?
A. order by acctno DESC;
B. order by 1;
C. order by sqrt(1);
D. order by acctno ASC;
12. You execute the query select 5 + 4 from DUAL. You have never
inserted data into the DUAL table before. Which of the following
statements best describes the DUAL table?
A. Dictionary view containing two schema names
B. Table with one column and one row used in various operations
C. Dictionary view containing two index names
D. Table with two columns and no rows used in various operations
13. You issue the following statement:
SELECT DECODE(ACCTNO, 123456, 'CLOSED', 654321, 'SEIZED',
590395, 'TRANSFER','ACTIVE') FROM BANK_ACCT;
If the value for ACCTNO is 503952, what information will this
statement display?
A. ACTIVE
B. TRANSFER
C. SEIZED
D. CLOSED
14. You are entering several dozen rows of data into the BANK_ACCT table.
Which of the following statements will enable you to execute the same
statement again and again, entering different values for variables at
statement runtime?
A. insert into BANK_ACCT (ACCTNO, NAME) VALUES
(123456,’SMITH’);
B. insert into BANK_ACCT (ACCTNO, NAME) VALUES (VAR1, VAR2);
ORACLE
Blind Folio 1:13
C. insert into BANK_ACCT (ACCTNO, NAME) VALUES (&VAR1,

’&VAR2’);
D. insert into BANK_ACCT (select ACCTNO, NAME from
EMP_BANK_ACCTS);
15. You execute the following SQL statement: select ADD_MONTHS
(‘28-APR-97’,120) from DUAL. What will Oracle return?
A. 28-APR-03
B. 28-APR-07
C. 28-APR-13
D. 28-APR-17
16. On Monday, June 26, 2037, at 10:30 at night, you issue the following
statement against an Oracle database:
ALTER SESSION SET NLS_DATE_FORMAT =
'DAY MONTH DD, YYYY: HH:MIAM';
Then, you issue the following statement
SELECT SYSDATE FROM DUAL;
What will Oracle return?
A. 26-JUN-37
B. June 26, 2037, 22:30
C. 26-JUN-2037
D. MONDAY JUNE 26, 2037: 10:30PM
17. You wish to join the data from two tables, A and B, into one result set and
display that set in your session. Tables A and B have a common column,
called C in both tables. Which of the following choices correctly displays
the where clause you would use if you wanted to see the data in table A
where the value in column C = 5, even when there was no corresponding
value in table B?
A. where A.C = 5 AND A.C = B.C;
B. where A.C = 5 AND A.C = B.C (+);
C. where A.C = 5 AND A.C (+) = B.C(+);
D. where A.C = 5;
ORACLE
Blind Folio 1:14
18. Each of the following statements is true about referential integrity, except
one. Which is it?
A. The referencing column in the child table must correspond with a
primary key in the parent.
B. All values in the referenced column in the parent table must be present
in the referencing column in the child.
C. The datatype of the referenced column in the parent table must be
identical to the referencing column in the child.
D. All values in the referencing column in the child table must be present
in the referenced column in the parent.
19. You have a group of values from a column in a table, and you would like
to perform a group operation on them. Each of the following functions
operate on data from all rows as a group, except for which of the
following choices?
A. avg( )
B. sqrt( )
C. count( )
D. stddev( )
20. You have a situation where you need to use the nvl( ) function. All
the following statements about the nvl( ) function are true except one.
Which is it?
A. nvl( ) returns the second value passed if the first value is NULL.
B. nvl( ) handles values of many different datatypes.
C. nvl( ) returns NULL if the first value is not equal to the second.
D. Both the values passed for nvl( ) must be the same datatype.
21. You are developing a stored procedure that handles table data. The
%rowtype expression in PL/SQL allows you to declare which of the
following kinds of variables?
A. Records
B. VARCHAR2s
ORACLE
Blind Folio 1:15
C. PLS_INTEGERs
D. NUMBERs
22. You create a sequence with the following statement:
CREATE SEQUENCE MY_SEQ
START WITH 394
INCREMENT BY 12
NOMINVALUE
NOMAXVALUE
NOCACHE
NOCYCLE;
Three users have already issued SQL statements to obtain NEXTVAL, and
four more have issued SQL statements to obtain CURRVAL. If you issue a
SQL statement to obtain the NEXTVAL, what will Oracle return?
A. 406
B. 418
C. 430
D. 442
23. Table EMP has 17,394,430 rows in it. You issue a delete from EMP
statement, followed by a commit. Then, you issue a select count(*)
to find out how many rows there are in the table. Several minutes later,
Oracle returns 0. Why did it take so long for Oracle to obtain this information?
A. The table was not empty.
B. The high-water mark was not reset.
C. Oracle always performs slowly after a commit is issued.
D. The table data did not exist to be counted anymore.
24. After creating a view, you realize that several columns were left out.
Which of the following statements would you issue in order to add some
columns to your view?
A. alter view
B. create or replace view
ORACLE
Blind Folio 1:16
C. insert into view

D. create view
25. You are testing several SQL statements for accuracy and usefulness. A SQL
statement will result in a Cartesian product as the result of which of the
following items?
A. A join statement without a where clause
B. The result of the sum( ) operation
C. select * from DUAL
D. The result of the avg( ) operation
26. In order to set your SQL*Plus session so that your NLS_DATE_FORMAT
information is altered in a specific way every time you log into Oracle,
what method would be used?
A. Setting preferences in the appropriate menu option.
B. Creating an appropriate login.sql file.
C. Issuing the alter user statement.
D. Issuing the alter table statement.
27. The EMP_SALARY table has two columns, EMP_USER and SALARY.
EMP_USER is set to be the same as the Oracle username. To support
user MARTHA, the salary administrator, you create a view with the
following statement:
CREATE VIEW EMP_SAL_VW
AS SELECT EMP_USER, SALARY
FROM EMP_SALARY
WHERE EMP_USER <> 'MARTHA';
MARTHA is supposed to be able to view and update anyone in the company’s
salary, except her own, through this view. Which of the following clauses
do you need to add to your view creation statement in order to implement
this functionality?
A. with admin option
B. with grant option
C. with security option
D. with check option
ORACLE
Blind Folio 1:17
28. You are developing PL/SQL code to manipulate and store data in an
Oracle table. All of the following numeric datatypes in PL/SQL can be
stored in an Oracle database, except one. Which is it?
A. CHAR
B. RAW
C. DATE
D. INTEGER
29. You are performing some conversion operations in your PL/SQL programs.
To convert a date value into a text string, you would use which of the
following conversion functions?
A. CONVERT
B. TO_CHAR
C. TO_NUMBER
D. TO_DATE
30. Your attempt to read the view creation code stored in the Oracle data
dictionary has encountered a problem. The view code appears to be
getting cut off at the end. In order to resolve this problem, which of the
following measures are appropriate?
A. Increase the size of the dictionary view.
B. Increase your user view allotment with the alter user statement.
C. Use the set long statement.
D. Use the set NLS_DATE_FORMAT statement.
31. You issue the following update statement against the Oracle database:
UPDATE BANK_ACCT SET NAME = 'SHAW';
Which records will be updated in that table?
A. The first record only
B. All records
C. The last record only
D. None of the records
ORACLE
Blind Folio 1:18
32. You are coding a complex PL/SQL block where several procedures call
other procedures. You have one outermost procedure that calls all other
procedures. If you only wanted to prevent the procedure from failing due
to a situation where a select into statement received two or more
records, you would include which of the following exceptions?
A. too_many_rows
B. others
C. zero_divide
D. no_data_found
33. You create a table but then subsequently realize you needed a few new
columns. To add those columns later, you would issue which of the
following statements?
A. create or replace table
B. alter table
C. create table
D. truncate table
34. You are busy creating your tables based on a logical data model. Which
of the following constraints require the references privilege in order
to be created?
A. UNIQUE
B. FOREIGN KEY
C. CHECK
D. NOT NULL
35. The INVENTORY table has three columns: UPC_CODE, UNITS, and
DELIV_DATE. The primary key is UPC_CODE. New records are added
daily through a view. The view was created using the following code:
CREATE VIEW DAY_INVENTORY_VW
AS SELECT UPC_CODE, UNITS, DELIV_DATE
FROM INVENTORY
WHERE DELIV_DATE = SYSDATE
WITH CHECK OPTION;
What happens when a user tries to insert a record with duplicate
UPC_CODE?
ORACLE
Blind Folio 1:19
A. The statement fails due to with check option clause.

B. The statement will succeed.
C. The statement fails due to PRIMARY KEY constraint.
D. The statement will insert everything except the date.
36. You need to search for text data in a column, but you only remember part
of the string. Which of the following SQL operations allows the use of
wildcard comparisons?
A. in
B. exists
C. between
D. like
37. You have a script you plan to run using SQL*Plus that contains one SQL
statement that inserts data into one table. Which of the following
options is the easiest way for this script to allow you to specify values for
variables once in the script, in a way where there is no user interaction?
A. Use define to capture value.
B. Use accept to capture value for each run.
C. Using & to specify values at runtime for the statement.
D. Hardcoded values in the statement.
38. You join data from two tables, EXPNS and EMP, into one result set and
display that set in your session. The tables have a common column called
EMPID. Which of the following choices correctly displays the where clause
you would use if you wanted to see the data in table EMP where the value in
column EMPID = 39284, but only when there is a corresponding value in
table EXPNS?
A. where EMP.EMPID = 39284 AND EMP.EMPID = EXPNS.EMPID;
B. where EMP.EMPID = 39284 (+) AND EMP.EMPID = EXPNS.EMPID;
C. where EMP.EMPID = EXPNS.EMPID;
D. where EMP.EMPID = 39284 AND EMP.EMPID = EXPNS.EMPID (+);
39. The %type expression in PL/SQL can be used to declare which of the
following kinds of variables?
ORACLE
Blind Folio 1:20
A. DATE variables
B. TEXT variables
C. PLS_INTEGER variables
D. REAL variables
40. Review the following transcript of a SQL*Plus session:
INSERT INTO INVENTORY (UPC_CODE, PRODUCT )
VALUES (503949353,'HAZELNUT COFFEE');
INSERT INTO INVENTORY (UPC_CODE, PRODUCT)
VALUES (593923506,'SKIM MILK');
INSERT INTO INVENTORY (UPC_CODE, PRODUCT)
VALUES (402392340,'CANDY BAR');
SAVEPOINT INV1;
UPDATE INVENTORY SET UPC_CODE = 50393950
WHERE UPC_CODE = 402392340;
UPDATE INVENTORY SET UPC_CODE = 4104930504
COMMIT;
UPDATE INVENTORY SET PRODUCT = (
SELECT PRODUCT FROM INVENTORY
WHERE UPC_CODE = 50393950)
ROLLBACK;
Which of the following UPC codes will not have records in the
INVENTORY table as a result of this series of operations?
A. 593923506
B. 503949353
C. 4104930504
D. 50393950
41. You are cleaning information out of the Oracle database. Which of the
following statements will get rid of all views that use a table at the same
time you eliminate the table from the database?
A. drop view
B. alter table
C. drop index
D. alter table drop constraint
ORACLE
Blind Folio 1:21
42. You want to join data from four tables into one result set and display that
set in your session. Table A has a column in common with table B, table B
with table C, and table C with table D. You want to further restrict data
returned from the tables by only returning data where values in the
common column shared by A and B equals 5. How many conditions
should you have in the where clause of your select statement?
A. 2
B. 3
C. 4
D. 5
43. You are developing some code in PL/SQL. If you wanted to declare
variables that could be used to store table column values, but didn’t
know the actual datatype of that column, PL/SQL allows you to declare
which of the following kinds of variables?
A. %rowtype variables
B. %type variables
C. FLOAT variables
D. VARCHAR2 variables
44. You are attempting to explain the Oracle security model for an Oracle
database to the new security administrator. What are two components
of the Oracle database security model?
A. Password authentication and granting privileges
B. Password authentication and creating database objects
C. Creating database objects and creating users
D. Creating users and password authentication
45. You have a script you plan to run using SQL*Plus that contains several
SQL statements that manage milk inventory in several different tables
based on various bits of information. You want the output to go into a
file for review later. Which command would you use?
A. prompt
B. echo
ORACLE
Blind Folio 1:22
C. spool
D. define
46. Your application’s business logic aligns closely with an Oracle internal
error. If you wanted to associate that internal error with a named exception
for handling in your application, in which of the following areas in your
procedure code must you include some support of this exception?
A. DECLARATION and EXCEPTION only.
B. DECLARATION, EXECUTION, and EXCEPTION.
C. EXCEPTION only.
D. No coding, definition, or exception handlers are required to raise
this exception.
47. You have a table called TEST_SCORE that stores test results by student
personal ID number, test location, and date the test was taken. Tests given
in various locations throughout the country are stored in this table. A
student is not allowed to take a test for 30 days after failing it the first
time, and there is a check in the application preventing the student from
taking a test twice in 30 days at the same location. Recently, it has come
to everyone’s attention that students are able to circumvent the 30-day
rule by taking a test in a different location. Which of the following SQL
statements would be useful for identifying the students who have done so?
A. select A.STUDENT_ID, A.LOCATION, B.LOCATION from
TEST_SCORE A, TEST_SCORE B where A.STUDENT_ID =
B.STUDENT_ID AND A.LOCATION = B.LOCATION AND
trunc(A.TEST_DATE)+30 <= trunc(B.TEST_DATE) AND
trunc(A.TEST_DATE)-30 >= trunc(B.TEST_DATE);
B. select A.STUDENT_ID, A.LOCATION, B.LOCATION from
B.STUDENT_ID AND A.LOCATION <> B.LOCATION AND
trunc(A.TEST_DATE)+30 >= trunc(B.TEST_DATE) AND
trunc(A.TEST_DATE)-30 <= trunc(B.TEST_DATE);
C. select A.STUDENT_ID, A.LOCATION, B.LOCATION from
B.STUDENT_ID AND A.LOCATION = B.LOCATION AND
ORACLE
Blind Folio 1:23
D. select A.STUDENT_ID, A.LOCATION, B.LOCATION from

trunc(A.TEST_DATE)+30 <= trunc(B.TEST_DATE) AND
trunc(A.TEST_DATE)-30 >= trunc(B.TEST_DATE);
48. In an expense application, you are searching for employee information in
the EMPLOYEE table corresponding to an invoice number you have. The
INVOICE table contains EMPID, the primary key for EMPLOYEE. Which
of the following options is appropriate for obtaining data from EMPLOYEE
using your invoice number?
A. select * from EMPLOYEE where empid = &empid;
B. select * from EMPLOYEE where empid = 69494;
C. select * from EMPLOYEE where empid = (select empid from
invoice where invoice_no = 4399485);
D. select * from EMPLOYEE;
49. Which of the following uses does not describe an appropriate use of the
having clause?
A. To put returned data into sorted order
B. To exclude certain data groups based on known criteria
C. To include certain data groups based on unknown criteria
D. To include certain data groups based on known criteria
50. You are managing data access for an application with 163 tables and
10,000 users. Which of the following objects would assist in managing
access in this application by grouping privileges into an object that can
be granted to users at once?
A. Sequences
B. Tables
C. Indexes
D. Roles
51. After logging on to Oracle the first time to access table EMP, user SNOW
is told to change his password. Which of the following statements allows
him to do so?
ORACLE
Blind Folio 1:24
A. alter user
B. alter table
C. alter role
D. alter index
52. User SNOW executes the following statement: select * from EMP.
This statement executes successfully, and SNOW can see the output.
Table EMP is owned by user REED. What object would be required in
order for this scenario to happen?
A. User SNOW would need the role to view table EMP.
B. User SNOW would need the privileges to view table EMP.
C. User SNOW would need a synonym for table EMP.
D. User SNOW would need the password for table EMP.
53. You develop a PL/SQL block containing a complex series of data changes.
A user then executes your PL/SQL block. At what point will the data
changes made be committed to the database?
A. When the PL/SQL block finishes
B. After each individual update
C. Whenever the commit command is issued
D. When you, the creator of the PL/SQL block, disconnect from your
session
54. If you would like to code your PL/SQL block to select some data from a
table, and then run through each row of output and perform some work,
which of the following choices best identifies how you would do so?
A. Implicit cursors with a cursor for loop
B. Implicit cursors with implicit cursor attributes
C. Explicit cursors with a cursor for loop
D. Explicit cursors with implicit cursor attributes
55. You have the following code block declaration in PL/SQL:
DECLARE
CURSOR EMP_1 IS
SELECT * FROM EMP
ORACLE
Blind Folio 1:25
WHERE EMPID = '40593';

CURSOR EMP_2 IS
SELECT * FROM EMP
WHERE EMPID = '50694';
BEGIN...
How could you rewrite this declaration block to reduce the number
of explicit cursors used in your program?
A. Using cursor for loops
B. Using %rowtype
C. Using %notfound
D. Passing EMPID values as parameters to the cursor
56. You issue the following statement in Oracle:
SELECT * FROM EMP WHERE DEPT IN
(SELECT DEPT FROM VALID_DEPTS
WHERE DEPT_HEAD = 'SALLY'
ORDER BY DEPT);
Which of the following choices best indicates how Oracle will respond
to this SQL statement?
A. Oracle returns the data selected.
B. Oracle returns data from EMP but not VALID_DEPTS.
C. Oracle returns data from VALID_DEPTS but not EMP.
D. Oracle returns an error.
57. You would like to reference a table in your PL/SQL block. What special
syntactic attribute must you precede the SQL statement with in order to
make the PL/SQL block compile?
A. Put a colon in front of all variables.
B. Use /* */ to surround the SQL code.
C. Prefix the command exec SQL in front of the statement.
D. No special syntax is required.
58. You are coding SQL statements in SQL*Plus. Which of the following is
a valid SQL statement?
A. select nvl(sqrt(59483)) from dual;
ORACLE
Blind Folio 1:26
B. select to_char(nvl(sqrt(59483), 0)) from dual;

C. select to_char(nvl(sqrt(59483), ’VALID’)) from dual;
D. select (to_char(nvl(sqrt(59483), ’0’)) from dual;
59. The following output is from a SQL*Plus session:
select PLAY_NAME||', ' || AUTHOR play_table from PLAYS;
My Plays and Authors
-------------------------------------
Midsummer Night's Dream, SHAKESPEARE
Waiting For Godot, BECKETT
The Glass Menagerie, WILLIAMS
Which of the following SQL*Plus commands produced it?
A. column PLAY_TABLE alias ”My Plays and Authors”
B. column PLAY_TABLE format a12
C. column PLAY_TABLE heading ”My Plays and Authors”
D. column PLAY_TABLE as ”My Plays and Authors”
60. You create a view with the following statement:
CREATE VIEW BASEBALL_TEAM_VW
AS SELECT B.JERSEY_NUM, B.POSITION, B.NAME
FROM BASEBALL_TEAM B
WHERE B.NAME = USER;
What will happen when user JONES attempts to select a listing for
user SMITH?
A. The select will receive an error.
B. The select will succeed.
C. The select will receive NO ROWS SELECTED.
D. The select will add data only to BASEBALL_TEAM.
ORACLE
Blind Folio 1:27
Practice Exam 2
1. Which of the following choices identifies a PL/SQL block containing the
correct syntax for a cursor for loop?
A. DECLARE
SELECT * FROM employee;
My_NameVARCHAR2(30);
BEGIN
OPEN My_Employees;
FOR csr_rec IN My_Employees LOOP
END LOOP;
CLOSE My_Employees;
END;
B. DECLARE
SELECT * FROM employee;
csr_rec VARCHAR2(30);
BEGIN
VALUES (csr_rec.name, csr_rec.title);
END LOOP;
END;
C. DECLARE
BEGIN
ORACLE
Blind Folio 1:28
END LOOP;
END;
D. DECLARE
My_Name VARCHAR2(30);
BEGIN
OPEN My_Employees;
LOOP
FETCH My_Employees INTO My_Name, My_Title;
END LOOP;
CLOSE My_Employees;
END;
2. Your attempt to read the view creation code stored in the Oracle data
dictionary has encountered a problem. The view code appears to be
getting cut off at the end. In order to resolve this problem, which of
the following measures are appropriate?
A. Increase the size of the dictionary view.
B. Increase your user view allotment with the alter user statement.
C. Use the set long statement.
D. Use the set NLS_DATE_FORMAT statement.
3. Inspect the following SQL statement:
SELECT FARM_NAME, COW_NAME,
COUNT(CARTON) AS NUMBER_OF_CARTONS
FROM COW_MILK
GROUP BY COW_NAME;
Which of the following choices contains the line with the error?
A. select FARM_NAME, COW_NAME,
B. count(CARTON) as NUMBER_OF_CARTONS
ORACLE
Blind Folio 1:29
C. from COW_MILK
D. group by COW_NAME;
E. There are no errors in the statement.
4. All of the following types of PL/SQL blocks are stored within the Oracle
database for reusability, except for one type. Which type is it?
A. Functions
B. Procedures
C. Package specs
D. Package bodies
E. Anonymous blocks
F. Triggers
5. Inspect the following SQL statement:
SELECT COW_NAME,
MOD(CARTON, FILL_STATUS)
FROM COW_MILK
GROUP BY COW_NAME;
Which of the following lines contains an error?
A. select COW_NAME,
B. mod(CARTON, FILL_STATUS)
C. from COW_MILK
D. group by COW_NAME;
E. There are no errors in this statement.
6. You are writing queries against an Oracle database. Which of the
following queries takes advantage of an inline view?
A. select * from EMP_VW where EMPID = (select EMPID from
INVOICE where INV_NUM = 5506934);
B. select A.LASTNAME, B.DEPT_NO from EMP A, (select EMPID,
DEPT_NO from DEPT) B where A.EMPID = B.EMPID;
ORACLE
Blind Folio 1:30
C. select * from EMP where EMPID IN (select EMPID from

INVOICE where INV_NUM > 23);
D. select ’select * from EMP_VW where EMPID is NOT NULL;’
from USER_TABLES;
7. For the following question, assume that before the following PL/SQL block
is executed, table MY_TAB contains one column called COLUMN1, and
one row with the value ‘FLIBBERJIBBER’.
DECLARE
VAR1 VARCHAR2(1);
VAR2 VARCHAR2(1);
IS
BEGIN
SELECT TO_CHAR(CEIL(SQRT(40)))
INTO VAR2
FROM DUAL;
SELECT SUBSTR(COLUMN1,4,1)
INTO VAR1
FROM MY_TAB;
IF VAR1 = 'J' THEN
VAR2 := '5';
ELSIF VAR2 = '7' THEN
VAR2 := 'L';
ELSE
VAR2 = '9';
END IF;
INSERT INTO MY_TAB VALUES (VAR2);
COMMIT;
END;
What is the value of COLUMN1 after executing this code block?
A. 5
B. 7
C. L
D. 9
E. J
ORACLE
Blind Folio 1:31
8. You create the following PL/SQL block:

DECLARE
VAR1 CONSTANT NUMBER := 90;
VAR2 NUMBER := 0;
BEGIN
SELECT ACCTNO
INTO VAR2
FROM BANK_ACCT
WHERE NAME = 'LEWIS';
VAR1 := VAR2 + 3049;
END;
Which of the following lines in this block of PL/SQL code will produce
an error?
A. VAR2 NUMBER := 0;
B. into VAR2
C. where NAME = ’LEWIS’;
D. VAR1 := VAR2 + 3049;
E. There are no errors in this PL/SQL block.
9. You are preparing to compile a block of PL/SQL code. The lines in the
block are shown in the following choices:
CREATE FUNCTION FOO (VAR1 IN VARCHAR2) IS
VAR2 VARCHAR2(1);
BEGIN
SELECT GENDER INTO VAR2 FROM EMP
WHERE LASTNAME = 'SMITHERS';
IF VAR1 = 6 THEN RETURN (6) ELSE RETURN (8);
END;
Which of the lines of PL/SQL code contain an error?
A. create function foo(VAR1 IN VARCHAR2) is
B. select GENDER into VAR2 FROM EMP
C. where LASTNAME = ’SMITHERS’;
ORACLE
Blind Folio 1:32
D. if VAR1 = 6 then RETURN (6) else return (8);

10. You have several indexes on a table that you want to remove. You want
to avoid the indexes associated with constraints, however. Each of the
following statements will work for removing an index associated with a
constraint, except one. Which choice is it?
A. drop index
B. alter table drop primary key cascade
C. alter table drop constraint
D. drop table
11. You are managing constraints on a table in Oracle. Which of the following
choices correctly identifies the limitations on CHECK constraints?
A. Values must be obtained from a lookup table.
B. Values must be part of a fixed set defined by create or alter table.
C. Values must include reserved words like sysdate and user.
D. Column cannot contain a NULL value.
12. Review the following statement:
CREATE TABLE FOOBAR
( MOO VARCHAR2(3),
BOO NUMBER);
This table contains 60,000,000 rows. You issue the following statement:
SELECT MOO, BOO FROM FOOBAR WHERE MOO = 'ABC'
This value appears in column MOO less than 10 percent of the time.
Yet, the query takes several minutes to resolve. Which of the following
explanations is the best reason why?
A. Oracle didn’t use the existing primary-key index.
B. select statements that do not use views take longer to resolve.
C. Table FOOBAR has no primary key, and therefore no index on MOO.
D. The table had been dropped and re-created.
ORACLE
Blind Folio 1:33
13. You have created a table called EMP with a primary key called
EMP_PK_01. In order to identify any objects that may be associated
with that table and primary key, what dictionary views and characteristics
would you look for?
A. USER_SEQUENCES, sequences created at the same time
B. USER_TABLES, tables with the same number of columns
C. USER_IND_COLUMNS, constraints with the same name as the table
D. USER_INDEXES, indexes with the same name as the constraint
14. You are designing your database, and you are attempting to determine the
best method for indexing your tables. Identify a main advantage for using
bitmap indexes on a database.
A. To improve performance on columns with many unique values
B. To improve performance on columns with few unique values
C. To improve performance on columns with all unique values
D. To improve performance on sequences with all unique values
DECLARE
CURSOR CARTON_CRSR IS
SELECT CARTON FROM MILK;
MY_CARTON MILK.CARTON%TYPE;
BEGIN
OPEN CARTON_CRSR;
LOOP
FETCH CARTON_CRSR INTO MY_CARTON;
INSERT INTO MY_MILK_CRATE (CARTON)
VALUES (MY_CARTON);
END LOOP;
CLOSE CARTON_CRSR;
END;
What is wrong with this PL/SQL block?
A. It will not work unless the loop is rewritten as a cursor for loop.
B. The exception handler must be defined if cursor is not declared.
C. The user does not have permission to execute the block.
ORACLE
Blind Folio 1:34
D. A loop exit condition must be defined.

E. There are no errors in this code block.
16. Your PL/SQL block includes the following statement:
SELECT EMP_ID
INTO MY_EMPID
FROM EMPLOYEE
WHEN LASTNAME = 'FRANKLIN';
You want Oracle to process the situation where no data is retrieved for
that LASTNAME value. Which of the following actions should be taken?
A. Include the %found implicit cursor attribute
B. Include the %notfound implicit cursor attribute
C. Include the when rowtype_mismatch exception
D. Nothing; Oracle raises this as an exception automatically.
17. User JANKO would like to insert a row into the EMPLOYEE table that has
three columns: EMPID, LASTNAME, and SALARY. The user would like to
enter data for EMPID 59694, LASTNAME Harris, but no salary. Which
statement would work best?
A. insert into EMPLOYEE values (59694,’HARRIS’, NULL);
B. insert into EMPLOYEE values (59694,’HARRIS’);
C. insert into EMPLOYEE (EMPID, LASTNAME, SALARY) values
(59694,’HARRIS’);
D. insert into EMPLOYEE (select 59694 from ’HARRIS’);
18. You join data from two tables, COW_MILK (C) and CARTON_CRATE
(C1), into one result set and display that set in your session. The tables
have a common column, called CARTON_NUM in both tables. Which of
the following choices correctly displays the where clause you would use
if you wanted to see the data in table COW_MILK for BESS the cow and
all corresponding information in CARTON_CRATE, where if there is no
data in CARTON_NUM, you don’t want to see the data in COW_MILK?
A. where C.COW_NAME <> ’BESS’ AND C.CARTON_NUM =
C1.CARTON_NUM;
B. where C.CARTON_NUM = C1.CARTON_NUM;
C. where C.COW_NAME = ’BESS’;
ORACLE
Blind Folio 1:35
D. where C.COW_NAME = ’BESS’ AND C.CARTON_NUM =

C1.CARTON_NUM;
E. where C.COW_NAME = ’BESS’ AND C.CARTON_NUM =
C1.CARTON_NUM (+);
19. You create a table with a primary key that is populated on insert with
a value from a sequence, then add several hundred rows to the table. You
then drop and re-create the sequence with the original sequence code.
Suddenly, your users are getting constraint violations. Which of the following
explanations is most likely the cause?
A. Dropping a sequence also removes any associated primary keys.
B. Any cached sequence values before it was dropped are unusable.
C. The table is read only.
D. The inserts contain duplicate data due to the reset sequence.
20. You are developing SQL statements for the application. Which of the
following SQL operations requires the use of a subquery?
A. in
B. exists
C. between
D. like
21. Review the following transcript from a SQL*Plus session:
SELECT CEIL(4093.505) FROM DUAL;
CEIL(4093.505)
--------------------
4094
Which single-row function could not be used to produce 4093 from the
number passed to the ceil( ) function?
A. round( )
B. trunc( )
C. floor( )
D. abs( )
22. You have a script you plan to run using SQL*Plus that contains several
SQL statements that update banking information for one person in several
ORACLE
Blind Folio 1:36
different tables based on name. Since the script only changes information
for one person, you want the ability to enter the name only once, and
have that information reused throughout the script. Which of the following
options is the best way to accomplish this goal in such a way that you
don’t have to modify the script each time you want to run it?
A. Use define to capture name value for each run.
B. Use accept to capture name value for each run.
C. Using the & character to specify lexical substitution for names at runtime.
D. Hardcode names in all SQL statements, and change the value each run.
23. You need to undo some data changes. Which of the following data
changes cannot be undone using the rollback command?
A. update
B. truncate
C. delete
D. insert
24. You are developing some code to handle transaction processing. Each of
the following items signifies the beginning of a new transaction, except
one. Which is it?
A. savepoint
B. set transaction
C. Opening a new session
D. commit
25. The following SQL statement is invalid:
SELECT PRODUCT, BRAND
WHERE UPC_CODE = '650-35365656-34453453454-45';
Which of the following choices indicate an area of change that would
make this statement valid?
A. A select clause
B. A from clause
C. A where clause
D. An order by clause
ORACLE
Blind Folio 1:37
DECLARE
CURSOR UPC_CODE_CRSR IS
SELECT UPC_CODE FROM INVENTORY;
MY_UPC_CODE INVENTORY%ROWTYPE;
BEGIN
OPEN UPC_CODE_CRSR;
LOOP
FETCH UPC_CODE_CRSR INTO MY_UPC_CODE;
EXIT WHEN UPC_CODE_CRSR%NOTFOUND;
INSERT INTO MY_SHOPPING_CART (UPC_CODE)
VALUES (UPC_CODE_CRSR);
END LOOP;
CLOSE UPC_CODE_CRSR;
END;
A. It will not work unless loop is rewritten as a cursor for loop.
B. The exception handler must be defined if a cursor is not declared.
C. The user does not have permission to execute the block.
D. Values from a cursor cannot be referenced directly.
27. You are coding a PL/SQL block. PROC_A calls PROC_B, which then calls
PROC_C, and PROC_B has no exception handler. If you wanted to prevent
the PROC_A procedure from failing due to a situation in PROC_B where
the divisor in a division statement was zero, how would you address this
in your code?
A. Use an if %zero_divide statement immediately following the
math operation.
B. Code a when zero_divide exception handler in PROC_C.
C. Code a when others exception handler in PROC_A.
D. Code a when others exception handler in PROC_C.
28. If you wanted to define an exception that caused no Oracle errors but
represented a violation of some business rule in your application, in which
of the following areas in your procedure code must you include some
support of this exception?
A. DECLARATION and EXCEPTION only
B. DECLARATION, EXECUTION, and EXCEPTION
ORACLE
Blind Folio 1:38
C. EXCEPTION only
D. No coding or definition is required to raise this exception.
29. You are at the beginning of your current transaction and want to prevent
your transaction from being able to change data in the database. To prevent
any statements in the current transaction from altering database tables,
which statement is used?
A. set transaction
B. rollback
C. commit
D. savepoint
30. Your application searches for data in the EMP table on the database on a
nullable column indicating whether a person is male or female. To improve
performance, you decide to index it. The table contains over 2,000,000
rows, and the column contains few NULL values. Which of the following
indexes would be most appropriate?
A. Nonunique B-tree index
B. Unique B-tree index
C. Bitmap index
D. Primary-key indexes
31. Your employee expense application stores information for invoices in one
table. Each invoice can have several items, which are stored in another
table. Each invoice may have one or more items, or none at all, but every
item must correspond to one invoice. The relationship between the INVOICE
table and INVOICE_ITEM table is best marked as what kind of relationship
on a logical data model?
A. Optional, one-to-many
B. Mandatory, one-to-many
C. Mandatory, one-to-one
D. Optional, one-to-one
ORACLE
Blind Folio 1:39
32. You issue the following statement:

SELECT DECODE(UPC_CODE, 40390, 'DISCONTINUED', 65421, 'STALE',
90395, 'BROKEN', 'ACTIVE') FROM INVENTORY;
If the value for ACCTNO is 20395, what information will this
statement display?
A. DISCONTINUED
B. STALE
C. BROKEN
D. ACTIVE
33. In considering the logical aspects to physical aspects of a database, which
of the following choices best represents the mapping of physical datafiles
to their logical counterparts?
A. Extents
B. Tablespaces
C. Segments
D. Blocks
34. You are comparing the storage implementation strategy employed by
Oracle to the strategy for storage implementation used in hierarchical
database systems like IMS. Which three of the following choices represent
advantages of the RDBMS implementation that are not present in hierarchical
databases? (Choose three.)
A. RDBMS requires that you define how to obtain data.
B. RDBMS defines how to obtain data for you.
C. RDBMS can model master/detail relationships.
D. RDBMS allows flexibility in changing data relationships.
E. RDBMS is able to model relationships other than master/detail.
35. Which of the following choices does not identify a benefit for using PL/SQL
that a developer might want to employ in an Oracle RDBMS application?
A. Ease of accessing data stored in Oracle
B. Ease of integrating programs written in different languages
ORACLE
Blind Folio 1:40
C. Ability to manipulate cursor data

D. Ability to handle errors without explicit conditional operations
36. You are developing advanced queries for an Oracle database. Which of
the following where clauses makes use of Oracle’s ability to logically test
values against a set of results returned without explicitly knowing what
the set is before executing the query?
A. where COL_A = 6
B. where COL_A in (6,7,8,9,10)
C. where COL_A between 6 AND 10
D. where COL_A in (select NUM from TAB_OF_NUMS)
37. You are developing a multiple-row query to handle a complex and
dynamic comparison operation in the Olympics. Two tables are involved.
CONTESTANT lists all contestants from every country, and MEDALS lists
every country and the number of gold, silver, and bronze medals they
have. If a country has not received one of the three types of medals, a
zero appears in the column. Thus, a query will always return data, even
for countries that haven’t won a medal. Which of the following queries
shows only the contestants from countries with more than 10 medallists
of any type?
A. select NAME from CONTESTANT C, MEDALS M where C.COUNTRY
= M.COUNTRY;
B. select NAME from CONTESTANT where COUNTRY C in (select
COUNTRY from MEDALS M where C.COUNTRY = M.COUNTY)
C. select NAME from CONTESTANT where COUNTRY C = (select
COUNTRY from MEDALS M where C.COUNTRY = M.COUNTY)
D. select NAME from CONTESTANT where COUNTRY in (select
COUNTRY from MEDALS where NUM_GOLD + NUM_SILVER +
NUM_BRONZE > 10)
38. You issue the following query in a SQL*Plus session:
SELECT NAME, AGE, COUNTRY FROM CONTESTANT
WHERE (COUNTRY, AGE) IN ( SELECT COUNTRY, MIN(AGE)
FROM CONTESTANT GROUP BY COUNTRY);
ORACLE
Blind Folio 1:41
Which of the following choices identifies both the type of query and the
expected result from the Oracle database?
A. Single-row subquery, the youngest contestant from one country
B. Multiple-row subquery, the youngest contestant from all countries
C. Multiple-column subquery, the youngest contestant from all countries
D. Multiple-column subquery, Oracle will return an error because = should
replace in
39. The contents of the CONTESTANTS table are listed as follows:
NAME AGE COUNTRY
---------------- -------------- ---------------
BERTRAND 24 FRANCE
GONZALEZ 29 SPAIN
HEINRICH 22 GERMANY
TAN 39 CHINA
SVENSKY 30 RUSSIA
SOO 21
You issue the following query against this table:
SELECT NAME FROM CONTESTANT
WHERE (COUNTRY, AGE) IN ( SELECT COUNTRY, MIN(AGE)
FROM CONTESTANT GROUP BY COUNTRY);
Which of the following contestants will not be listed among the output?
A. SOO
B. HEINRICH
C. BERTRAND
D. GONZALEZ
40. An object in Oracle contains many columns that are functionally
dependent on the key column for that object. The object requires
segments to be stored in areas of the database other than the data
dictionary. The object in question is correctly referred to as which
of the following objects?
A. CURSOR
B. TABLE
ORACLE
Blind Folio 1:42
C. SEQUENCE
D. VIEW
41. You are defining where to place information in a PL/SQL block. If you
wanted the block to contain a conditional operation that determined
whether a portion of code would be executed based on the value returned
from a query, which section of the PL/SQL block would you write the
code into?
A. Declaration
B. Executable
C. Exception
D. Package specification
42. You can use the following PL/SQL block to answer this question:
DECLARE
CURSOR UPC_CODE_CRSR IS
SELECT UPC_CODE FROM INVENTORY;
BEGIN
MY_UPC_CODE INVENTORY%ROWTYPE;
OPEN UPC_CODE_CRSR;
LOOP
FETCH UPC_CODE_CRSR INTO MY_UPC_CODE;
EXIT WHEN UPC_CODE_CRSR%NOTFOUND;
INSERT INTO MY_SHOPPING_CART (UPC_CODE)
VALUES (UPC_CODE_CRSR);
END LOOP;
CLOSE UPC_CODE_CRSR;
END;
What is wrong with this code block?
A. The variable is declared incorrectly.
B. Appropriate looping values are not used.
C. The insert statement is incorrectly defined.
D. The loop must be closed in the exception handler.
43. You need to execute a PL/SQL procedure. Which of the following choices
does not represent a way to do so?
A. With the execute command from SQL*Plus
B. From within a procedure
ORACLE
Blind Folio 1:43
C. From within a select statement

D. Using the start command
44. You are declaring variables in PL/SQL. Which two of the following choices
are not methods that can be used to define variables with an initial value?
(Choose two.)
A. VAR_1 CONSTANT NUMBER(3) :=96;
B. VAR_1 NUMBER(3) :=96;
C. VAR_1 NUMBER(3) DEFAULT 96;
D. VAR_1 NUMBER(3) INITIAL 96;
45. You are developing a PL/SQL block in Oracle. Which of the following
standard parts of a PL/SQL block are required for proper compilation
and use?
A. Declaration section
B. Exception section
C. Execution section
D. Parameter passing
46. You may use the following PL/SQL block to answer this question:
DECLARE
PROCEDURE UPDATE_EMP_SAL ( P_EMPID IN NUMBER, P_SAL IN NUMBER)
IS BEGIN
UPDATE EMP SET SALARY = SAL WHERE EMPID = P_EMPID;
END;
MY_EMPID NUMBER := 12345;
MY_SAL NUMBER DEFAULT 50000;
BEGIN
UPDATE_EMP_SAL (MY_EMPID, MY_SAL);
END;
A. Blocks containing named subprograms must also be named.
B. Variables passed as parameters must have the same name as the
parameter.
C. The nested block must be defined after variables used in the main block.
D. The default keyword is invalid in PL/SQL.
ORACLE
Blind Folio 1:44
47. Your application uses database objects in distributed systems owned by

many users. In masking complexity added by referencing those various
objects, which of the following choices identify a situation where use of
synonyms is not permitted?
A. Package constants in the local procedure
B. Procedures in a remote system
C. Procedures owned by another user
D. Tables owned by another user in a remote system
48. You are developing a complex PL/SQL routine. Which of the following
control structures in PL/SQL can be used for unconditional branching to
a particular area of the code block?
A. if true then
B. loop
C. if-then-else
D. goto
NAME AGE COUNTRY
---------------- -------------- ---------------
BERTRAND 24 FRANCE
GONZALEZ 29 SPAIN
HEINRICH 22 GERMANY
TAN 39 CHINA
SVENSKY 30 RUSSIA
SOO 21
Use the following PL/SQL block to answer the question:
DECLARE
VAR1 VARCHAR2(20);
VAR2 NUMBER := 0;
BEGIN
SELECT NAME INTO VAR1 FROM CONTESTANT
WHERE COUNTRY = 'JAPAN';
IF VAR1 = 'GERMANY' THEN
VAR2 := LENGTH(VAR1);
ELSIF VAR1 = 'RUSSIA' THEN
VAR2 := LENGTH(VAR1) + 15;
ELSIF VAR1 = 'FRANCE' THEN
ORACLE
Blind Folio 1:45

ELSIF VAR1 = 'GERMANY' THEN
ELSIF VAR1 = 'CHINA' THEN
ELSE
END IF;
END;
What is the value of VAR2 at the end of this code block?
A. 7
B. 21
C. 55
D. NULL
NAME AGE COUNTRY
---------------- -------------- ---------------
BERTRAND 24 FRANCE
GONZALEZ 29 SPAIN
HEINRICH 22 GERMANY
TAN 39 CHINA
SVENSKY 30 RUSSIA
SOO 21
You may use the following PL/SQL block to answer this question:
DECLARE
JUNK VARCHAR2(20);
i PLS_INTEGER := 0;
BEGIN
SELECT NAME INTO JUNK FROM CONTESTANT
WHERE COUNTRY = 'RUSSIA';
FOR i IN 1..50 LOOP
IF JUNK = 'SVENSKY' THEN
GOTO END_LOOP;
END IF;
JUNK := 'RIZENFRANZ';
<<end_loop>>
END LOOP;
END;
ORACLE
Blind Folio 1:46
What is wrong with this statement?

A. The label is not followed by an executable statement.
B. The endless loop must be modified.
C. The looping variable must be assigned a default value.
D. There is nothing wrong with this statement.
51. You want to create a user-defined record corresponding to the columns
in table CONTESTANT. The columns in this table include:
NAME – VARCHAR2(20)
AGE – NUMBER(5)
COUNTRY – VARCHAR2(30)
Which of the following code fragments handle this declaration properly?
A. CONTESTANT_RECORD_NAME VARCHAR2(20);
CONTESTANT_RECORD_AGE NUMBER(5);
CONTESTANT_RECORD_COUNTRY VARCHAR2(30);
B. TYPE cont_rec IS (
NAME VARCHAR2(20),
AGE NUMBER(5),
COUNTRY VARCHAR2(30));
CONTESTANT_RECORD cont_rec;
C. CONTESTANT_RECORD CONTESTANT%ROWTYPE;
D. CONTESTANT_RECORD RECORD OF TABLE CONTESTANT;
52. You are loading data into the CONTESTANTS table, which contains the
following columns:
NAME – VARCHAR2(20)
AGE – NUMBER(5)
COUNTRY – VARCHAR2(30)
To use a PL/SQL table for doing so, which of the following code fragments
identifies how you would declare it?
A. CONTESTANT_TABLE is table of VARCHAR2(20), NUMBER(5),
VARCHAR2(30) index by BINARY_INTEGER;
B. type CONT_REC is (
NAME VARCHAR2(20),
AGE NUMBER(5),
ORACLE
Blind Folio 1:47
CONTESTANT_TABLE is table of CONT_REC%TYPE index by

BINARY_INTEGER;
C. CONTESTANT_TABLE is table of CONTESTANT%ROWTYPE;
D. CONTESTANT_TABLE is table of CONTESTANT%ROWTYPE index
by BINARY_INTEGER;
53. Examine the following code block:
DECLARE
CONTESTANT_NAME_TABLE IS TABLE OF
CONTESTANT.NAME%TYPE INDEX BY BINARY_INTEGER;
CURSOR CONTESTANT_CSR IS SELECT NAME
FROM CONTESTANT;
MYVAL BINARY_INTEGER := 0;
BEGIN
FOR MY_CONT_REC IN CONTESTANT_CSR LOOP
CONTESTANT_NAME_TABLE(MYVAL) := MY_CONT_REC.NAME;
MYVAL := MYVAL + 1;
END LOOP;
–– HOW MANY ROWS IN TABLE?
END;
If you want to find out the number of rows in the table, what two methods
can you use to do so in the fewest number of program lines possible?
(Choose two.)
A. select count(1) from CONTESTANT_NAME_TABLE;
B. Value in MYVAL
C. CONTESTANT_NAME_TABLE.COUNT
D. Use another loop to cycle through the records in CONTESTANT_TABLE
54. You are developing a PL/SQL block designed for bulk data operations.
When attempting to store table data for multiple rows in a PL/SQL
variable, which of the following choices identifies the mechanism best
suited for the task?
A. Record
B. Cursor
C. PL/SQL table
D. PL/SQL table of records
ORACLE
Blind Folio 1:48
55. You declare a cursor as follows: cursor c1 is select * from

CONTESTANT for update. Which of the following choices correctly
describes what happens when this cursor is opened?
A. Oracle returns an error because the way the records should be updated
is not defined.
B. Oracle retrieves all records in the CONTESTANT table only.
C. Oracle retrieves all records in the CONTESTANT table into the cursor
and locks those rows for update.
D. Oracle updates all records in the CONTESTANT table as specified in the
open statement.
56. You develop a PL/SQL block that will be used in an overall application
that records grants for political asylum for Olympic athletes to the country
where the games are held. The cursor definition in this block is cursors
asy_csr is select NAME from CONTESTANT. As you loop through the
cursor, you want the ability to update the record back in the table that
corresponds to the record in the cursor. Which of the following choices
best identifies the way to do so without explicit comparison between the
NAME value in the cursor and the NAME value in the table?
A. update CONTESTANT set COUNTRY = ’USA’ where current of
asy_csr;
B. Redefine the cursor as cursor cnt_csr is select NAME from
CONTESTANT for update;
C. update CONTESTANT set COUNTRY = ’USA’ where NAME =
asy_csr.NAME;
D. You cannot achieve this functionality with PL/SQL.
57. You want to define a cursor that contains a subquery. Which of the following
choices best identifies how PL/SQL allows you to handle this situation?
A. PL/SQL does not allow cursors to contain subqueries.
B. PL/SQL only allows cursors to contain subqueries resolving to inline
views.
ORACLE
Blind Folio 1:49
C. PL/SQL only allows cursors to contain subqueries resolving to a set of

comparison values for a where clause.
D. PL/SQL allows cursors to contain the same types of subqueries permitted
in any SQL query.
58. You develop a PL/SQL program with several blocks three layers deep.
PROC_A calls PROC_B, which in turn calls PROC_C. Which of the
following choices correctly identifies how Oracle will propagate the
error in order to find an exception handler that will resolve an exception
being raised in the innermost block?
A. PROC_B, PROC_A, PROC_C
B. PROC_C, PROC_B, PROC_A
C. PROC_A, PROC_B, PROC_C
D. Exception is propagated directly to the user level
59. You are developing PL/SQL process flow into your program. The command
used to open a cursor for loop is which of the following keywords?
A. open
B. fetch
C. parse
D. None, cursor for loops handle cursor opening implicitly.
60. You are determining the appropriate program flow for your PL/SQL
application. Which of the following statements are true about while loops?
A. Explicit exit statements are required in while loops.
B. Counter variables are required in while loops.
C. An if-then statement is needed to signal when a while loop should
end.
D. All exit conditions for while loops are handled in the exit when
clause.
ORACLE
Blind Folio 1:50
Practice Exam 3
1. You are implementing transaction control in a PL/SQL block. Which of
the following choices best describes the use of autonomous transactions
in your database?
A. A developer wants records to be added to an EXIT_STATUS table
regardless of whether the transaction completed successfully or not.
B. Two users on the Oracle database are modifying records in different
tables at the same time.
C. Two users on the Oracle database are modifying different records in the
same table at the same time.
D. Two users on the Oracle database are modifying the same records in the
same table at the same time.
2. You need to compute an N-dimensional cross-tabulation in your SQL
statement output for reporting purposes. Which of the following Oracle8i
clauses can be used for this purpose?
A. having
B. cube
C. rollup
D. trim( )
3. You are using summaries in the Oracle8i database for data warehousing.
Which of the following clauses is implicitly added by Oracle8i to queries
in order to take advantage of summary management?
A. order by
B. group by
C. cube
D. rollup
4. You are managing a data warehouse using Oracle8i. Which of the following
choices identifies a key aspect of dimension creation that must be addressed
when the dimension is created?
A. cube operation
B. rollup operation
ORACLE
Blind Folio 1:51
C. Summary management
D. Level hierarchy
5. You need to define a collection type for student exam records in an
Oracle8i database. Each record will consist of student vital information,
along with the answers each student gave on every exam they took in the
class. You want the ability to reference specific questions on exams for
individual students for comparison purposes. Which of the following
choices identifies the best way to do it?
A. Scalar datatypes
B. User-defined type
C. VARRAY
D. Nested table
6. You are indexing Oracle data in an application. The index will be on
a column containing sequential numbers with at least seven significant
digits. Most, if not all, entries will start with 1. Which of the following
indexes would be best suited for the task?
A. B-tree indexes
B. Reverse-key indexes
C. Bitmap indexes
D. Descending indexes
7. You need to store a large block of text data in Oracle. These text blocks
will be around 3,500 characters in length. Which datatype would you use
for storing these large objects?
A. VARCHAR2
B. CLOB
C. BLOB
D. BFILE
8. Two users exist on an Oracle8i system, named FLUFFY and MUFFY.
MUFFY owns a PL/SQL procedure called foobar( ), defined as follows:
create procedure FOOBAR
authid current_user
is
ORACLE
Blind Folio 1:52
var1 number;
begin
select * into var1 from foo;
end;
To use procedure foobar( ), what privileges does FLUFFY need granted to
her (choose as many as appropriate)?
A. create procedure on schema MUFFY
B. execute on procedure foobar( )
C. update on table FOO
D. select on table FOO
9. User MILTON issues the following code block in his SQL*Plus session:
CREATE OR REPLACE PROCEDURE upd_tran
( PVAL1 IN VARCHAR2) IS
PRAGMA AUTONOMOUS_TRANSACTION;
MY_VAL2 NUMBER(10);
BEGIN
SELECT COL_1 INTO MY_VAL2
FROM TAB_2 WHERE COL_2 = PVAL1;
UPDATE TAB_1 SET COL_1 = PVAL1
WHERE COL_2 = PVAL2;
COMMIT;
END;
Later, another procedure called my_tran( ) is defined by user GOETHE,
which calls upd_tran( ) as part of an application. Which of the following
statements is true about this application?
A. Any transaction in progress in my_tran( ) when upd_tran( ) is
called will be committed when upd_tran( ) completes.
B. Any uncommitted changes made to table TAB_2 by my_tran( ) will
not be seen by upd_tran( ).
C. Procedure upd_tran( ) has no problem updating TAB_1 if my_tran(
) has already done so without issuing a commit
before the call to upd_tran( ).
D. User GOETHE needs to have update privileges on TAB_1 to execute
procedure upd_tran( ).
ORACLE
Blind Folio 1:53
10. Developer JANET receives an error due to the following statement in the
DECLARATION section:
PI CONSTANT NUMBER;
The problem is because of which of the following causes?
A. There is not enough memory in the program for the constant.
B. There is no value associated with the constant.
C. There is no datatype associated with the constant.
D. pi is a reserved word.
11. You are designing your PL/SQL exception handler inside a nested block.
Which statement most accurately describes the result of not creating an
exception handler for a raised exception?
A. The program will continue without raising the exception.
B. There will be a memory leak.
C. Control will pass to the PL/SQL block caller’s exception handler.
D. The program will return a %notfound error.
12. You are determining what types of cursors to use in your PL/SQL code.
Which of the following statements is true about implicit cursors?
A. Implicit cursors are used for SQL statements that are not named.
B. Developers should use implicit cursors with great care.
C. Implicit cursors are used in cursor for loops to handle data
processing.
D. Implicit cursors are no longer a feature in Oracle.
13. You are constructing PL/SQL process flow for your program. Which of
the following is not a feature of a cursor for loop?
A. Record-type declaration
B. Opening and parsing of SQL statements
C. Fetches records from cursor
D. Requires exit condition to be defined
ORACLE
Blind Folio 1:54
14. A developer would like to use a referential datatype declaration on

a variable. The variable name is EMPLOYEE_LASTNAME, and the
corresponding table and column is EMPLOYEE and LASTNAME,
respectively. How would the developer define this variable using
referential datatypes?
A. Use EMPLOYEE.LASTNAME%type.
B. Use EMPLOYEE.LASTNAME%rowtype.
C. Look up datatype for EMPLOYEE column on LASTNAME table and
use that.
D. Declare it to be type LONG.
15. After executing an update statement, the developer codes a PL/SQL
block to perform an operation based on SQL%rowcount. What data is
returned by the SQL%rowcount operation?
A. A Boolean value representing the success or failure of the update
B. A numeric value representing the number of rows updated
C. A VARCHAR2 value identifying the name of the table updated
D. A LONG value containing all data from the table
16. You are defining a check following a SQL statement to verify that the
statement returned appropriate data. Which three of the following are
implicit cursor attributes? (Choose three.)
A. %found
B. %too_many_rows
C. %notfound
D. %rowcount
E. %rowtype
17. You are constructing PL/SQL process flow into your program. If left out,
which of the following would cause an infinite loop to occur in a simple loop?
A. loop
B. end loop
C. if-then
D. exit
ORACLE
Blind Folio 1:55
18. You are coding your exception handler. The others exception handler is
used to handle all of the following exceptions, except one. Which exception
does the others exception handler not cover?
A. no_data_found
B. others
C. rowtype_mismatch
D. too_many_rows
19. You are defining a cursor in your PL/SQL block. Which line in the
following statement will produce an error?
A. cursor action_cursor is
B. select name, rate, action
C. into action_record
D. from action_table;
E. There are no errors in this statement.
20. You are developing PL/SQL process flow into your program. The command
used to open a cursor for loop is which of the following keywords?
A. open
B. fetch
C. parse
D. None, cursor for loops handle cursor opening implicitly.
21. You are determining the appropriate program flow for your PL/SQL
application. Which of the following statements are true about while loops?
A. Explicit exit statements are required in while loops.
B. Counter variables are required in while loops.
C. An if-then statement is needed to signal when a while loop
should end.
D. All exit conditions for while loops are handled in the exit
when clause.
ORACLE
Blind Folio 1:56
22. For the following question, assume that before the following PL/SQL block
is executed, table OPERATIONS contains one column called STATUS, and
one row with the value PROCESSING.
DECLARE
VAR1 VARCHAR2(1);
VAR2 VARCHAR2(1);
IS
BEGIN
SELECT TO_CHAR(CEIL(SQRT(40)))
INTO VAR2
FROM DUAL;
SELECT SUBSTR(STATUS,4,1)
INTO VAR1
FROM OPERATIONS;
IF VAR1 = 'H' THEN
VAR2 := '5';
ELSIF VAR1 = 'C' THEN
VAR2 := 'L';
ELSE
VAR2 = '9';
END IF;
INSERT INTO OPERATIONS VALUES (VAR2);
COMMIT;
END;
What is the value of the STATUS column after executing this code block?
A. 5
B. 7
C. L
D. 9
E. J
23. You create the following PL/SQL block:
DECLARE
MYVAR1 CONSTANT NUMBER := 100;
MYVAR2 NUMBER := 0;
BEGIN
SELECT ITEM
INTO VAR2
FROM ITEM_TABLE
WHERE NAME = 'SMITH';
ORACLE
Blind Folio 1:57
VAR1 := VAR2 + 3079;

END;
Which of the following lines in this block of PL/SQL code will produce
an error?
A. VAR2 NUMBER := 0;
B. into VAR2
C. where NAME = ’SMITH’;
D. VAR1 := VAR2 + 3079;
24. You are preparing to compile a block of PL/SQL code. The lines in the
block are shown in the choices below:
CREATE FUNCTION OPERATION_A (VAR1 IN VARCHAR2) RETURN NUMBER IS
VAR2 VARCHAR2(1);
BEGIN
SELECT GENDER INTO VAR1 FROM EMP
WHERE LASTNAME = 'SMITHERS';
IF VAR1 = 6 THEN RETURN (6) ELSE RETURN (8);
END IF;
END;
Which of the lines of PL/SQL code contain an error?
A. create function foo(VAR1 IN VARCHAR2) return NUMBER is
B. select GENDER into VAR1 from EMP
C. where LASTNAME = ’SMITHERS’;
D. if VAR1 = 6 then return (6) else return (8);
25. Dropping a table has which of the following effects on a nonunique index
created for the table?
A. No effect.
B. The index will be dropped.
C. The index will be rendered invalid.
D. The index will contain NULL values.
ORACLE
Blind Folio 1:58
26. Which of the following statements about indexes is true?

A. Columns with low cardinality are handled well by B-tree indexes.
B. Columns with low cardinality are handled poorly by bitmap indexes.
C. Columns with high cardinality are handled well by B-tree indexes.
27. To add the number of columns selected by a view:
A. Add more columns to the underlying table.
B. Issue the alter view statement.
C. Use a correlated subquery in conjunction with the view.
D. Drop and re-create the view with references to select more columns.
28. Which of the following choices are valid parameters for sequence creation?
A. identified by
B. using temporary tablespace
C. maxvalue
D. on delete cascade
29. The following statement is issued against the Oracle database. Which line
will produce an error?
A. create view EMP_VIEW_01
B. as select E.EMPID, E.LASTNAME, E.FIRSTNAME, A.ADDRESS
C. from EMPLOYEE E, EMPL_ADDRESS A
D. where E.EMPID = A.EMPID
E. with check option;
F. This statement contains no errors.
30. You are granting privileges on your table to another user. Which object
privilege allows the user to create his or her own table with a foreign key
on a column in your table?
A. references
B. index
C. select
D. delete
ORACLE
Blind Folio 1:59
31. Which of the following statements are true about roles? (Choose three.)
A. Roles can be granted to other roles.
B. Privileges can be granted to roles.
C. Roles can be granted to users.
D. Roles can be granted to synonyms.
32. After referencing NEXTVAL, the value in CURRVAL:
A. Is incremented by one
B. Is now in PREVVAL
C. Is equal to NEXTVAL
D. Is unchanged
33. The EMP_SALARY table has two columns, EMP_USER and SALARY.
EMP_USER is set to be the same as the Oracle username. To support
user MARTHA, the salary administrator, you create a view with the
following statement:
CREATE VIEW EMP_SAL_VW
AS SELECT EMP_USER, SALARY
FROM EMP_SALARY
WHERE EMP_USER <> 'MARTHA';
MARTHA is supposed to be able to view and update anyone’s salary in
the company, except her own, through this view. Which of the following
clauses do you need to add to your view creation statement in order to
implement this functionality?
A. with admin option
B. with grant option
C. with security option
D. with check option
34. The INVENTORY table has three columns: UPC_CODE, UNITS, and
DELIV_DATE. The primary key is UPC_CODE. New records are added
daily through a view. The view was created using the following code:
CREATE VIEW DAY_INVENTORY_VW
AS SELECT UPC_CODE, UNITS, DELIV_DATE
FROM INVENTORY
WHERE DELIV_DATE = SYSDATE
WITH CHECK OPTION;
ORACLE
Blind Folio 1:60
What happens when a user tries to insert a record with duplicate

UPC_CODE?
A. The statement fails due to with check option clause.
B. The statement will succeed.
C. The statement fails due to PRIMARY KEY constraint.
D. The statement will insert everything except the date.
35. You are cleaning information out of the Oracle database. Which of the
following statements will get rid of all views that use a table at the same
time you eliminate the table from the database?
A. drop view
B. alter table
C. drop index
D. alter table drop constraint
36. You create a view with the following statement:
CREATE VIEW BASEBALL_TEAM_VW
AS SELECT B.JERSEY_NUM, B.POSITION, B.NAME
FROM BASEBALL_TEAM B
WHERE B.NAME = USER;
What will happen when user JONES attempts to select a listing for
user SMITH?
A. The select will receive an error.
B. The select will succeed.
C. The select will receive NO ROWS SELECTED.
D. The select will add data only to BASEBALL_TEAM.
37. Which of the following integrity constraints automatically create an index
when defined? (Choose two.)
A. Foreign keys
B. UNIQUE constraints
C. NOT NULL constraints
D. Primary keys
ORACLE
Blind Folio 1:61
38. Which of the following dictionary views gives information about the
position of a column in a primary key?
A. ALL_PRIMARY_KEYS
B. USER_CONSTRAINTS
C. ALL_IND_COLUMNS
D. ALL_TABLES
39. Developer ANJU executes the following statement: create table
ANIMALS as select * from MASTER.ANIMALS; What is the effect
of this statement?
A. A table named ANIMALS will be created in the MASTER schema with
the same data as the ANIMALS table owned by ANJU.
B. A table named ANJU will be created in the ANIMALS schema with the
same data as the ANIMALS table owned by MASTER.
C. A table named ANIMALS will be created in the ANJU schema with the
D. A table named MASTER will be created in the ANIMALS schema with
the same data as the ANJU table owned by ANIMALS.
40. User JANKO would like to insert a row into the EMPLOYEE table that has
three columns: EMPID, LASTNAME, and SALARY. The user would like to
enter data for EMPID 59694, LASTNAME Harris, but no salary. Which
statement would work best?
A. insert into EMPLOYEE values (59694,’HARRIS’, NULL);
B. insert into EMPLOYEE values (59694,’HARRIS’);
C. insert into EMPLOYEE (EMPID, LASTNAME, SALARY) values
(59694,’HARRIS’);
D. insert into EMPLOYEE (select 59694 from ’HARRIS’);
41. No relationship officially exists between two tables. Which of the
following choices is the strongest indicator of a parent/child relationship?
A. Two tables in the database are named VOUCHER and
VOUCHER_ITEM, respectively.
B. Two tables in the database are named EMPLOYEE and PRODUCTS,
respectively.
ORACLE
Blind Folio 1:62
C. Two tables in the database were created on the same day.

D. Two tables in the database contain none of the same columns.
42. Which of the following are valid database table datatypes in Oracle?
(Choose three.)
A. CHAR
B. VARCHAR2
C. BOOLEAN
D. NUMBER
43. Omitting the where clause from a delete statement has which of the
following effects?
A. The delete statement will fail because there are no records to delete.
B. The delete statement will prompt the user to enter criteria for the
deletion.
C. The delete statement will fail because of syntax error.
D. The delete statement will remove all records from the table.
44. Which line of the following statement will produce an error?
A. create table GOODS
B. (GOODNO number,
C. GOOD_NAME varchar2(20) check(GOOD_NAME in (select NAME
from AVAIL_GOODS)),
D. constraint PK_GOODS_01
E. primary key (GOODNO));
F. There are no errors in this statement.
45. The transaction control that prevents more than one user from updating
data in a table is which of the following?
A. Locks
B. Commits
C. Rollbacks
D. Savepoints
ORACLE
Blind Folio 1:63
46. To increase the number of nullable columns for a table:

A. Use the alter table statement.
B. Ensure that all column values are NULL for all rows.
C. First, increase the size of adjacent column datatypes, and then add
the column.
D. Add the column, populate the column, and then add the NOT NULL
constraint.
47. A user issues the statement select count(*) from EMPLOYEE. The
query takes an inordinately long time and returns a count of zero. The
most cost-effective solution is to:
A. Upgrade the hardware.
B. Truncate the table.
C. Upgrade the version of Oracle.
D. Delete the high-water mark.
48. You are creating some tables in your database as part of the logical data
model. Which of the following constraints have an index associated with
them that is generated automatically by Oracle?
A. UNIQUE
B. FOREIGN KEY
C. CHECK
D. NOT NULL
49. Each of the following statements is true about referential integrity, except
one. Which is it?
A. The referencing column in the child table must correspond with a
primary key in the parent.
B. All values in the referenced column in the parent table must be present
C. The datatype of the referenced column in the parent table must be
identical to the referencing column in the child.
D. All values in the referencing column in the child table must be present
in the referenced column in the parent.
ORACLE
Blind Folio 1:64
50. You are managing constraints on a table in Oracle. Which of the following
choices correctly identifies the limitations on CHECK constraints?
A. Values must be obtained from a lookup table.
B. Values must be part of a fixed set defined by create or alter table.
C. Values must include reserved words like SYSDATE and USER.
D. Column cannot contain a NULL value.
51. Which of the following is not a group function?
A. avg( )
B. sqrt( )
C. sum( )
D. max( )
52. In order to perform an inner join, which criteria must be true?
A. The common columns in the join do not need to have shared values.
B. The tables in the join need to have common columns.
C. The common columns in the join may or may not have shared values.
D. The common columns in the join must have shared values.
53. Once defined, how long will a variable remain defined in SQL*Plus?
A. Until the database is shut down
B. Until the instance is shut down
C. Until the statement completes
D. Until the session completes
54. You want to change the prompt Oracle uses to obtain input from a user.
Which two of the following choices are used for this purpose? (Choose two.)
A. Change the prompt in the config.ora file.
B. Alter the prompt clause of the accept command.
C. Enter a new prompt in the login.sql file.
D. There is no way to change a prompt in Oracle.
ORACLE
Blind Folio 1:65
55. No search criteria for the EMPLOYEE table are known. Which of the
following options is appropriate for use when search criteria are unknown
for comparison operations in a select statement? (Choose two.)
A. select * from EMPLOYEE where empid = &empid;
B. select * from EMPLOYEE where empid = 69494;
C. select * from EMPLOYEE where empid =
(select empid from invoice where invoice_no = 4399485);
D. select * from EMPLOYEE;
56. The default character for specifying substitution variables in select
statements is
A. Ampersand
B. Ellipses
C. Quotation marks
D. Asterisk
57. A user is setting up a join operation between tables EMPLOYEE and DEPT.
There are some employees in the EMPLOYEE table that the user wants
returned by the query, but the employees are not assigned to department
heads yet. Which select statement is most appropriate for this user?
A. select e.empid, d.head from EMPLOYEE e, dept d;
B. select e.empid, d.head from EMPLOYEE e, dept d where
e.dept# = d.dept#;
C. select e.empid, d.head from EMPLOYEE e, dept d where
e.dept# = d.dept# (+);
D. select e.empid, d.head from EMPLOYEE e, dept d where
e.dept# (+) = d.dept#;
58. Which three of the following uses of the having clause are appropriate?
(Choose three.)
A. To put returned data into sorted order
B. To exclude certain data groups based on known criteria
C. To include certain data groups based on unknown criteria
D. To include certain data groups based on known criteria
ORACLE
Blind Folio 1:66
59. A Cartesian product is

A. A group function
B. Produced as a result of a join select statement with no where clause
C. The result of fuzzy logic
D. A special feature of Oracle server
60. The default character that identifies runtime variables is changed by:
A. Modifying the initsid.ora file
B. Modifying the login.sql file
C. Issuing the define variablename command
D. Issuing the set define command
ORACLE
Blind Folio 1:67
Answers to Practice Exam 1

1. A. select NAME, JERSEY_NO where JERSEY_NO = 6;
Explanation SQL statements in Oracle must have a from clause. A SQL statement
can lack a where clause, in which case all data in the table will be returned. However,
if the statement does not have a from clause, Oracle will not know what table to
retrieve data from. Recall that a special table called DUAL assists in situations where
you don’t want to retrieve data from a table, but instead want only to manipulate
expressions. (Topic 2.2)
2. D. cursor for loop

Explanation Any time you see a PL/SQL block where the cursor is opened and
manipulated manually, you can revise the code into a cursor for loop to take
advantage of Oracle automatically managing many aspects of cursor manipulation
for you. The other choices simply rewrite the existing structure, and may save a line
or two of coding. But, the cursor for loop will tremendously reduce your coding
burden in this situation. (Topic 22.3)
3. B. others
Explanation A special exception called others exists in PL/SQL that will handle
all unhandled exceptions captured at that level of processing. Although Oracle will
raise the zero_divide exception automatically whenever you attempt to divide a
number by zero, coding an exception handler for zero_divide only defines what
happens when zero_divide is raised. It does not, for example, handle situations
where another exception gets raised, as others does. The no_data_found and
too_many_rows exceptions are raised when no data or too much data is returned
by an implicit or explicit cursor, respectively. (Topic 24.4)
4. C. raise_application_error
Explanation The raise_application_error function allows Oracle to return
a user-defined error message when an exception condition is raised. The pragma
exception_init keywords (of which pragma is a subset) indicate a compiler
directive where you tell Oracle you want to associate an internal error with an
exception name of your own devises. The others exception is a catchall for processing
exceptions that would otherwise escape unhandled. Finally, the exception keyword
is used to denote the exceptions section of your code block. (Topic 24.6)
ORACLE
Blind Folio 1:68
5. A. Issuing an update statement

Explanation The only choice that does not end a transaction is the one that
continues the transaction, namely issuing another update statement. A commit
tells Oracle to save your data changes and end the transaction, while a rollback
tells Oracle to discard your data changes and end the transaction. Closing SQL*Plus
or otherwise ending the session is usually treated as an implicit commit and ends
your transaction as well. (Topic 10.3)
6. B. commit
Explanation In order to save any change you make in Oracle, you will use
the commit command. The savepoint command merely identifies a logical
breakpoint in your transaction that you can use to break up complex units of work.
The rollback command discards every change you made since the last commit.
Finally, the set transaction command will set up the transaction to be
read-only against the Oracle database. (Topic 10.5)
7. D. USER_CONS_COLUMNS
Explanation The USER_CONS_COLUMNS dictionary view shows you all the
columns in tables belonging to that user that are part of indexes used to enforce
constraints. USER_INDEXES is incorrect because that view only displays information
about the index itself, not the columns in the index. USER_TAB_COLUMNS displays
all the columns in all tables owned by that user. Finally, USER_COLUMNS is not an
actual view in the Oracle database. (Topic 14.2)
8. A. UNIQUE
Explanation Only UNIQUE and PRIMARY KEY constraints require Oracle to
generate an index that supports or enforces the uniqueness of the column values.
Foreign keys do not require this sort of index. CHECK constraints also do not require
an index. Finally, NOT NULL constraints do not require an index, either. (Topic 12.1)
9. D. The view
Explanation When you drop a table with the cascade constraints option,
Oracle removes all associated indexes, triggers, and constraints that reference that
table from other tables. Oracle does not remove the views that use that table,
however—you must do that manually with the drop view statement. (Topic 11.5)
ORACLE
Blind Folio 1:69
10. D. MONTHS_BETWEEN
Explanation Each of the choices accepts as input a DATE datatype and returns a
DATE datatype—with one exception. The MONTHS_BETWEEN function returns a
number indicating how many months there are between the two dates you give it.
This number will be displayed with numbers to the right of the decimal point, which
you can round off if you like. All the rest of the choices return a DATE value in one
form or another. (Topic 4.2)
11. C. order by sqrt(1);

Explanation Although the order by clause allows you to refer to the column you
want order determined by either by the column name or by the number representing
the column order in the select clause of the select statement, you cannot
perform any sort of numeric function on that number. Both the asc and desc
keywords are valid for the order by clause, indicating ascending order (default)
and descending order, respectively. (Topic 3.2)
12. B. Table with one column and one row used in various operations
Explanation The DUAL table is a special table in Oracle used to satisfy the
requirement of a from clause in your SQL statements. It contains one column and
one row of data. It is not a dictionary view; rather, it is an actual table. Thus, you
should understand why the answer is what it is in this question. As a side note, you
would use the DUAL table in arithmetic expressions and would not actually pull
real data from the database. Finally, you should never insert data into the DUAL
table under any circumstance. (Topic 2.1)
13. A. ACTIVE
Explanation The decode( ) function is used as a “case” statement, where Oracle
will nest through the value in the column identified in the first parameter (in this
case ACCTNO). If that value equals the second parameter, the third parameter is
returned. If that value equals the fourth parameter, the fifth parameter will be returned,
and so on. If the value equals no parameter, the default value provided in the last
parameter (in this case ACTIVE) is returned. TRANSFER would be returned if
ACCTNO=590395, SEIZED would be returned if ACCTNO=654321, and CLOSED
would be returned if ACCTNO=123456. (Topic 4.1)
ORACLE
Blind Folio 1:70
14. C. insert into BANK_ACCT (ACCTNO, NAME) VALUES (&VAR1,

’&VAR2’);
Explanation In order to have statement reusability where you can enter a value on
the fly, you must use lexical references as runtime variables. These references are
preceded with an ampersand (&) character, as in the correct answer. Though you
can use nested subqueries in your insert statements, this has the effect of inserting
multiple rows at once without requiring input from the user. (Topic 9.1)
15. B. 28-APR-07
Explanation On this question, you really have to put your thinking cap on.
ADD_MONTHS adds a specified number of months indicated by the second
parameter to the value in the first parameter; 120 months is 10 years, so if you add
10 to the year in the date given you should come up with 28-APR-07, which is the
correct answer. Important Point: beware of having too much of your time sucked
up by this sort of “brain teaser” question. (Topic 4.2)
16. D. MONDAY JUNE 26, 2037: 10:30PM

Explanation The first statement in this question alters the date format shown in
your SQL*Plus session, while the second statement returns the current date and time
in that specific format. In this case, your format is the day of the week, followed by
the month of the year, the date, the year, and the time in AM/PM format. This being
the case, the correct answer is MONDAY JUNE 26, 2037: 10:30PM. (Topic 9.2)
17. B. where A.C = 5 AND A.C = B.C (+);

Explanation The correct choice illustrates your use of Oracle’s outer join function.
The question indicates you want to see data in table A, no matter whether there is
corresponding data in table B or not. Thus, you place the outer join operation next
to the reference to the C column in table B. If the outer join operation is removed
(it looks like a (+)), Oracle will only return data from table A, for which there are
corresponding records in Table B. If the outer join operator is used for both tables,
you will get a syntax error. If you omit the join operator comparing values from
table A to table B, Oracle will return a Cartesian product of the data you requested
from A with all the data from table B. (Topic 5.2)
18. B. All values in the referenced column in the parent table must be present
Explanation Referential integrity is from child to parent, not vice versa. The parent
table can have many values that are not present in child records, but the child
ORACLE
Blind Folio 1:71
record must correspond to something in the parent. Thus, the correct answer is all
values in the referenced column in the parent table must be present in the
referencing column in the child. (Topic 12.1)
19. B. sqrt( )
Explanation All the choices indicate group by functions, except for the sqrt( )
function, which is a single-row function acting on each value in each column row,
one at a time or individually. avg( ) processes data from multiple rows in a
column and produces one result, the average value for all of them. count( )
processes all values in a column or columns and counts the number of row values
in that column or columns. The stddev( ) function takes all values in a column
of rows and determines the standard deviation for that set of values. (Topic 6.1)
20. C. nvl( ) returns NULL if the first value is not equal to the second.
Explanation The only statement in this question that is not true is nvl( ) returns
NULL if the first value is not equal to the second. nvl( ) is specifically designed to
avoid returning NULL for a column, by substituting another value that you pass as
the second parameter. All other statements made in these choices are true about
nvl( )—it handles many different datatypes, and both values passed must be the
same datatype. (Topic 4.1)
21. A. Records
Explanation %rowtype is a special keyword in PL/SQL that allows you to define a
record that conforms to the datatypes for rows in a particular table, as in EMP%rowtype
for a set of datatypes in a row from the EMP table. Although that row may contain
columns of VARCHAR2 or NUMBER datatype, a record datatype is a more accurate
way to describe this feature. However, because PLS_INTEGER data cannot be stored
in Oracle tables, you will never see this datatype in a %rowtype record. (Topic 21.2)
22. B. 418
Explanation The key here is being able to distinguish between CURRVAL and
NEXTVAL. Only NEXTVAL will actually change the sequence value. CURRVAL only
selects the current value, so you can factor out the four people who have issued
statements requesting CURRVAL and pay attention only to those users requesting
NEXTVAL. There are two of those, so the sequence, which started at 394, has been
incremented twice by 12. 394 + 12 +12 = 418. (Topic 15.2)
ORACLE
Blind Folio 1:72
23. B. The high-water mark was not reset.

Explanation The select count(*) statement takes a long time because Oracle had
to inspect the table in its entirety in order to derive the rowcount, even though the table
was empty. To avoid this situation on large tables, use the truncate statement rather
than delete. truncate resets the high-water mark on your table, thus reducing the
time it takes Oracle to perform select count(*) operations. (Topic 11.5)
24. B. create or replace view

Explanation The column definitions for a view can only be changed by re-creating
the view with the create or replace view statement. The alter view command
is used only to recompile a view, while insert into view is not a valid SQL
statement. Although create view will technically work, you must first drop the
view you want to re-create, which requires two statements, not one. create or
replace view is the most accurately correct choice offered. (Topic 13.2)
25. A. A join statement without a where clause

Explanation Cartesian products are the result of select statements that contain
malformed where clauses. sum( ) and avg( ) operations are group functions and
do not produce Cartesian products. Selecting data from the DUAL table will not
produce a Cartesian product because only one table is involved—and a table with
only one row at that! (Topic 5.1)
26. B. Creating an appropriate login.sql file

Explanation SQL*Plus shows its roots in UNIX systems through the login.sql
file. This file is used to specify settings used in your session. login.sql runs
automatically after you log in to Oracle. SQL*Plus in Windows environments does
not have a Preferences menu, eliminating that choice. You shouldn’t attempt to use
the alter table or alter user statements for this purpose, either. (Topic 9.5)
27. D. with check option

Explanation The appropriate clause is with check option. You can add this
clause to a create view statement so that the view will not allow you to add
rows to the underlying table that cannot then be selected in the view. The with
admin/grant option clauses are used to assign administrative ability to users
along with granting them a privilege. The with security option is a work of
fiction—it does not exist in Oracle. (Topic 13.3)
ORACLE
Blind Folio 1:73
28. D. INTEGER
Explanation Although you can declare variables in PL/SQL blocks using the INTEGER
datatype, you cannot store INTEGER datatype data in Oracle tables. All other datatypes
shown, namely CHAR, RAW, and DATE, can all be stored in the Oracle database, as
well as being used as datatypes for PL/SQL variables. (Topic 11.3)
29. B. TO_CHAR
Explanation TO_CHAR is used to convert DATE values, numbers, and other things
into text strings. The CONVERT operation is used to convert a text string from one
character set to another, while the TO_NUMBER operation converts numeric text to
true numbers. The TO_DATE function is used to convert a properly formatted text
string into a DATE value. (Topic 4.3)
30. C. Use the set long statement.

Explanation The set long command allows you to increase the buffer size
SQL*Plus will use to retrieve LONG data values. This statement is used because
the view text is stored in a LONG column in the appropriate dictionary view. The
dictionary view itself does not need to be increased, nor can you somehow solve
this problem with the alter user statement. Finally, the NLS_DATE_FORMAT
option is used for date formatting, not LONG columns. (Topic 9.2)
31. B. All records

Explanation Since the update statement does not contain a where clause, the
change will be made to every record in the table. There is no way to accurately
update only the first or last record in the table. No records will be updated only
if there is something wrong with the update statement, such as a column being
referenced incorrectly. (Topic 10.1)
32. A. too_many_rows
Explanation The answer to this question is too_many_rows, an Oracle predefined
exception that gets raised automatically in the situation described in the question.
The others exception is a catchall that handles any exception that would otherwise
go unhandled at this level of execution. zero_divide is raised automatically
whenever you attempt to divide by 0. Finally, no_data_found is the conceptual
opposite of the correct answer. (Topic 24.3)
ORACLE
Blind Folio 1:74
33. B. alter table

Explanation The alter table statement allows you to easily add columns after
the table is created, with minimal impact to your system. Unlike views, you do not
use the or replace keyword for this effort, thus creating a powerful distractor for
the user who is more familiar with views than with underlying tables. The create
table could be used for the task, but you would first need to issue the drop
table statement to get rid of the initial table. (Topic 11.4)
34. B. FOREIGN KEY

Explanation Foreign-key relationships require that you grant references privileges
on a table to the user creating the foreign-key relationship from their table to yours.
There is no particular special privilege that must be granted to create unique, check,
or NOT NULL constraints other than create table. (Topic 16.3)
35. C. The statement fails due to PRIMARY KEY constraint.

Explanation It should be obvious that the statement fails—the real question here
is why. The reason is because of the PRIMARY KEY constraint on UPC_CODE. As
soon as you try to add a duplicate record, the table will reject the addition. Although
the view has with check option specified, this is not the reason the addition
fails. It would be the reason an insert fails if you attempt to add a record for a
day other than today, however. (Topic 13.4)
36. D. like
Explanation In the situation where you want to use wildcards, Oracle offers the
like comparison operator. This operator allows you to search for text strings like
the one you’re looking for. The in operator specifies a set of values to which the
comparison value can be equal to one of, while exists allows you to use a
subquery as a lookup validity test for some piece of information. between specifies
a range comparison, such as between 1 AND 5. (Topic 3.1)
37. A. Use define to capture value.

Explanation The define command can be used to identify a variable and assign
it a value for use throughout a script running in SQL*Plus. This is useful when
executing a number of SQL statements in batch. Although the accept command
can perform the same function, the key factor that makes this the wrong answer is
mention of no user interaction in the question. Hardcoded values will work, but
they make the script almost completely not reusable. Finally, although lexical
references using ampersand (&) followed by a label will allow you statement
reusability, your users will have to keep entering values every time a statement
containing the lexical reference is processed. (Topic 9.4)
ORACLE
Blind Folio 1:75
38. A. where EMP.EMPID = 39284 AND EMP.EMPID = EXPNS.EMPID;

Explanation Since you only want data from either table where there is a match in
the other, you are performing a regular join or “equijoin” operation. In Oracle, you
would not use the outer join (+) operator for this purpose. This eliminates all the
choices where an outer join operator is displayed. (Topic 5.1)
39. A. DATE variables

Explanation %type can be used to declare a variable as the same datatype as a
table column, as in EMPLOYEE.LASTNAME%type for the LASTNAME column of
the EMPLOYEE table. All other datatypes offered as choices are valid datatypes in
PL/SQL, but not in Oracle tables, so you could not use the %type keyword to
reference these datatypes in your code. Of the choices given, only DATE variables
can be used as column datatypes in Oracle. (Topic 19.2)
40. C. 4104930504
Explanation The only record that will not be present from the choices given is
4104930504, because UPC code #402392340 does not exist at the time this
statement is issued. It was already changed to 50393950, and thus the 4104930504
update fails when you issue it. As an aside, this question is really tricky because in
order to get the answer right you have to read the question for a long time, and that
wastes time when you’re taking the OCP exams. Important Note: This question will
take up an enormous amount of time if you’re not careful. (Topic 10.5)
41. A. drop view

Explanation When a table is dropped, Oracle eliminates all related database
objects, such as triggers, constraints, and indexes—except for views. Views are
actually considered separate objects, and although the view will not function
properly after you drop the underlying table, Oracle will keep the view around
after the table is dropped. (Topic 13.5)
42. C. 4
Explanation The general rule of thumb here is that, if you have n tables you want
to join, four in this case, you will generally need n − 1 comparison operations in your
where clause joined together by AND—three in this case. In addition, recall from
the question that you want to restrict return data further based on values in the first
ORACLE
Blind Folio 1:76
table. Thus, your where clause would have four conditions, and may look something
like the following block:
WHERE
A.COLUMN1 = 5 AND
A.COLUMN1 = B.COLUMN1 AND
B.COLUMN2 = C.COLUMN2 AND
C.COLUMN3 = D.COLUMN3
(Topic 5.1)
43. B. %type variables

Explanation You would use the %type keyword to assign dynamically the datatype
of a variable according to the datatype in a column of a table. %rowtype is used
for declaring variable records according to all columns in a table, so it is almost, but
not quite, correct. VARCHAR2 variables might be the result of using %type, but
remember, you didn’t actually know what the datatype of the variable you were
declaring was—only that it was the same for a particular column of a table. Finally,
you will never wind up with a FLOAT variable, because Oracle does not allow
FLOAT variables to be used as datatypes in table columns. (Topic 19.2)
44. A. Password authentication and granting privileges

Explanation Although in order to get database access you need to create user
privileges, the two real components of Oracle’s security model are password
authentication and granting privileges. When a user is created, the user will still
not be able to connect to Oracle unless they are granted a privilege (create
session), and even when they connect they still cannot see anything unless
someone gives them permission to via the grant command. (Topic 16.1)
45. C. spool
Explanation The spool command makes SQL*Plus write an output file containing
all information transacted in the session, from the time you turn spooling on and
identify the output file to the time you either turn spooling off or end the session.
prompt causes SQL*Plus to prompt you to enter data using a specialized request
message of your own devising, while echo causes an error because it is not a
valid command in SQL*Plus. Finally, the define command is used for variable
definition and variable assignment in SQL*Plus scripts. (Topic 2.3)
46. A. DECLARATION and EXCEPTION, only.

Explanation The DECLARATION section of your code block must contain the
pragma keyword, a compiler directive used to associate named exceptions with
ORACLE
Blind Folio 1:77
internal Oracle errors. The exception handler must contain an exception handler
for the named exception. The only area that does not require coding is the execution
section of your code block, because Oracle will automatically raise your named
exception whenever the internal error occurs in code execution. (Topic 24.1)
47. B. select A.STUDENT_ID, A.LOCATION, B.LOCATION from

Explanation This question is probably the hardest on the exam. Even if you have
a bit of SQL experience, this question will take you a while. When taking OCP, the
last thing you need is time-waster questions to throw you off. By ensuring that the
student is the same and that the date the test was taken violated the 30-day rule,
and that the test location is not the same, you correctly select choice B. A good
technique to avoid having questions like this one consume all your time is to skip
it if you cannot answer it within 30 seconds. You’ll most likely have some time at
the end of the exam to review the questions you skipped. (Topic 5.3)
48. C. select * from EMPLOYEE where empid = (select empid from

invoice where invoice_no = 4399485);
Explanation If you can use a subquery, you should do so. There is only one choice
that displays a subquery, so that one must be the answer. All the other choices depend
on the EMPID being provided, not the invoice number. (Topic 7.1)
49. A. To put returned data into sorted order

Explanation The having clause is best used to include or exclude certain data
groups, not to return data in sort order. The order by clause handles that task.
(Topic 6.4)
50. D. Roles
Explanation Roles allow you to group privileges together into one object and grant
the privileges to the user at one time. There are no privileges related to indexes
other than the privilege to access the associated table. Tables and sequences both
require privileges to be granted to a user or role; they do not simplify the act of
privilege management in any way. (Topic 16.2)
Friday, February 09, 2001 3:04:48 PM
ORACLE
Blind Folio 1:78
51. A. alter user

Explanation The alter user statement with the identified by clause is used
to change a user’s password. alter role is used for modifying the actual role
object, and affects users insofar as the user has been granted the role. Of the remaining
choices, although user SNOW may be able to execute those statements depending
on what privileges he is granted, none of these privileges will handle what the
question requires. (Topic 16.1)
52. C. User SNOW would need a synonym for table EMP.

Explanation User SNOW needs a synonym in order to refer to a table he doesn’t
own without prefixing that reference with a schema owner. Without privileges, SNOW
would not see the data, but even with the appropriate privileges granted, SNOW
still needs to prefix the table name with the schema information if no synonym
exists for the table in that schema. If there is no synonym, SNOW still has to prefix
references to EMP with REED, as in REED.EMP. Tables don’t have passwords,
databases do, so that choice is patently incorrect. (Topic 15.4)
53. C. Whenever the commit command is issued

Explanation In PL/SQL, data changes are committed using the commit command,
and will only take place when that command is issued. Oracle will not automatically
commit data changes when the PL/SQL block ends, eliminating that choice. Only
if a commit command is present after each individual update will the change be
committed. Finally, if you are the creator of the block, when you disconnect from
Oracle has nothing to do with another user making changes to data with your code.
(Topic 19.4)
54. C. Explicit cursors with a cursor for loop

Explanation Explicit cursors with a cursor for loop is the way to go for this
situation. You would first need to write an explicit cursor so that you could work
with each row individually, which eliminates all the implicit cursor choices. There
is no way to “write” an implicit cursor, as implicit cursors are handled by the Oracle
RDBMS in a way that is transparent to the user. However, the second part of the
question is how you will move through each row of output. Here, you need a loop
more than you need cursor attributes. (Topic 22.1)
55. D. Passing EMPID values as parameters to the cursor

Explanation You would reduce the number of cursors declared by having a single
cursor that accepted a value for EMPID as a parameter. Although you might use the
cursor in a cursor for loop, you wouldn’t rely on that feature to reduce the overall
number of cursors in your database. The variable declaration keyword %rowtype
ORACLE
Blind Folio 1:79
helps you reduce the amount of work you need to do to set up a record variable,
but does nothing to reduce the number of cursors you declare. Finally, %notfound
is an implicit cursor attribute that again has nothing to do with how many cursors
you have to declare in your code block. (Topic 23.1)
56. D. Oracle returns an error.

Explanation In this situation, you cannot use the order by clause in a subquery.
Oracle will return an error. Thus, no data will be returned from any table, so all
other choices are wrong. (Topic 7.4)
57. D. No special syntax is required.

Explanation No special syntax is required to place SQL statements into your
PL/SQL block. The /* */ markers are used for commenting out portions of your
code, while exec SQL and colons are required only when embedding SQL
statements into Pro*C programs, a topic not covered by this exam. (Topic 19.3)
58. B. select to_char(nvl(sqrt(59483), 0)) from dual;

Explanation The select to_char(nvl(sqrt(59483), 0)) from dual;
statement is a valid statement. The select nvl(sqrt(59483)) from dual;
statement does not pass enough parameters to the nvl( ) function. The select
TO_CHAR(nvl(sqrt(59483), ’VALID’)) from dual; statement breaks the rule
in nvl( ) that states that both parameters passed into the function must be the same
datatype. The select (to_char(nvl(sqrt(59483), ’0’)) from dual;
statement is missing a matching closing parenthesis after ‘0’. (Topic 4.2)
59. C. column PLAY_TABLE heading ”My Plays and Authors”

Explanation The heading clause to the column command in SQL*Plus acts in
the same way as a column alias does in SQL—it modifies the output of the query to
use a heading of your design. Despite its similarity, however, the heading clause is
not the same as an alias in SQL. Thus, both the choice identifying the alias clause
and the choice using the as keyword are both incorrect. The choice containing the
format clause should be easy to eliminate. (Topic 9.2)
60. C. The select will receive NO ROWS SELECTED.

Explanation Although the query will succeed (translation—you won’t receive an
error), you must beware of the distracter in choice B. In reality, choice C is the better
answer because it more accurately identifies what really will occur when you issue
this statement. This view will behave as any select statement would when you list
ORACLE
Blind Folio 1:80
criteria in the where clause that no data satisfies, by returning NO ROWS SELECTED.
This is not an error condition, but you wouldn’t call it a successful search for data,
either, making both those choices incorrect. Finally, select statements never add
data to a table. (Topic 13.3)

1. C.
DECLARE
BEGIN
END LOOP;
END;
Explanation If you understand the nature of a cursor for loop, you should be
able to easily identify the correct answer. Otherwise, you should skip this question
and come back to it because you will wind up wasting a lot of time reading the
code for each choice—time better spent answering other easier questions. For your
information, a cursor for loop does three things for you. It declares the fetch
variable (csr_rec in the code block) implicitly. It also handles the exit condition
automatically. Finally, it takes care of opening and closing the cursor without
requiring you to code explicit open and close commands. (Topic 23.3)
2. C. Use the set long statement.

Explanation The set long command allows you to increase the buffer size
SQL*Plus will use to retrieve LONG data values. This statement is used because
the view text is stored in a LONG column in the appropriate dictionary view. The
dictionary view itself does not need to be increased, nor can you somehow solve
this problem with the alter user statement. Finally, the NLS_DATE_FORMAT
option is used for date formatting, not LONG columns. (Topic 13.1)
3. D. group by COW_NAME;
Explanation The problem with this statement is that not enough leading columns
from the query are referred to in the group by clause. As a result, you will receive
the ORA-00979 (not a group by expression) error. The correct group by clause
ORACLE
Blind Folio 1:81
would read group by FARM_NAME, COW_NAME;. All other areas of the statement
are syntactically and semantically correct. (Topic 6.3)
4. E. Anonymous blocks
Explanation All PL/SQL blocks are blocks that Oracle can store within the database,
except for anonymous blocks. These are compiled and run when you submit them
to Oracle, and then eventually discarded. There is little opportunity for you to reuse the
code unless you store the code as a text file and rerun it in your session. (Topic 17.1)
5. B. mod(CARTON, FILL_STATUS)
Explanation The line containing reference to the mod( ) operation is the one
containing the error. Because this is a single-row function, it cannot be used as the
group by expression in a SQL statement containing the group by expression. The
rest of the statement is correct. If you substituted a grouping expression like sum( ),
avg( ), or count( ), you would have a correct statement. (Topic 6.2)
6. B. select A.LASTNAME, B.DEPT_NO from EMP A, (select

EMPID, DEPT_NO FROM DEPT) B where A.EMPID = B.EMPID;
Explanation An inline view is an undeclared view consisting only of a parenthetical
select statement in a from clause. This subquery is then treated like a view in
other areas of the main query. While choices A and C involve use of subqueries,
only the use of a subquery in the from clause of a select statement constitutes
an inline view. Choice D is not really a subquery at all, but rather a way to get
SQL*Plus to write a select statement for every line of output from the query.
(Topic 8.3)
7. C. L
Explanation The square root of 40 is a fraction between 6 and 7, which rounds up
to 7 according to the algorithm behind the ceil( ) function. This means that the
VAR2 = ‘7’ flag in the elsif will resolve to true. Thus, VAR2 is set to ‘L’, and then
written to the database with the insert statement at the end. Be careful not to waste
time on reviewing all the intricacies of the PL/SQL block provided. (Topic 20.2)
8. D. VAR1 := VAR2 + 3049;

Explanation The main problem with this block of PL/SQL code has to do with the
VAR1 := VAR2 + 3049 statement. This is because VAR1 cannot be assigned a value
in this code block because the variable is defined as a constant. VAR2 NUMBER :=
0; is a proper variable declaration. The into VAR2 clause is appropriate in a
ORACLE
Blind Folio 1:82
PL/SQL fetch statement. Finally, the where NAME = ’LEWIS’; clause is well
constructed. All other lines of code in the block not identified as choices are
syntactically and semantically correct. (Topic 17.4)
9. A. create function foo(VAR1 IN VARCHAR2) is

Explanation There is no definition of return value datatype in this code block,
making the function declaration line the correct answer. Although it may seem that
the if-then statement in the third line of the code block is incorrect because you
are comparing a VARCHAR2 variable to the number “6”, in reality Oracle handles
this situation just fine because there is an implicit type conversion occurring in the
background. Finally, the select, into, from, and where clauses of the fetch
statement are all constructed correctly. (Topic 17.2)
10. D. drop table

Explanation This is not an easy question. To rid your table of an index associated
with a constraint, you will need to do something to the table or constraint, not to
the index. The alter table drop constraint works for indexes associated with unique
constraints, and the alter table drop primary key statement works for indexes
associated with primary keys. Dropping the table removes all associated database
objects, including indexes. However, simply dropping the index with the drop
index statement is not possible. (Topic 15.3)
11. B. Values must be part of a fixed set defined by create or alter table.
Explanation A CHECK constraint may only use fixed expressions defined when
you create or alter the table with the constraint definition. The reserved words like
SYSDATE and USER, or values from a lookup table are not permitted, making those
answers incorrect. Finally, NULL values in a column are constrained by NOT NULL
constraints, a relatively unsophisticated form of check constraints. (Topic 12.2)
12. C. Table FOOBAR has no primary key, and therefore no index on MOO.
Explanation Because table FOOBAR has no primary key, you cannot obtain data
from it rapidly the way you could if the MOO column was set up as the primary
key, and thus indexed. So, although Oracle not using the primary key is technically
true, the more accurate answer is that the table had no primary key to use, and
therefore no index on MOO. There are no views involved, and the table didn’t
get dropped. (Topic 11.2)
ORACLE
Blind Folio 1:83
13. D. USER_INDEXES, indexes with the same name as the constraint

Explanation Since an associated index is generated automatically by Oracle, you
will look in the USER_INDEXES view to find the associated object. And, since Oracle
uses the name of the constraint in order to name the index created with that constraint,
you will look for indexes that have the name EMP_PK_01. There is little correlation
between tables and sequences, other than sequences can be used to populate the
columns of a table. The USER_TABLES view will offer little value either, since only
the EMP table will appear in it, and you are not looking for another table with the
same name. (Topic 14.1)
14. B. To improve performance on columns with few unique values

Explanation Bitmap indexes are primarily designed to improve performance on
searches that involve column data that is static and has few unique values. The
ideal example of this is whether a person is male or female—it usually doesn’t
change, and these are the only two choices. All other choices referring to table
columns in this question identify when it is appropriate to use B-tree indexes, the
other main type of index available in Oracle. Also, understand that indexes do not
improve performance on sequences. (Topic 15.1)
15. D. A loop exit condition must be defined.

Explanation No loop exit condition is defined for this code block. You don’t need
an exception handler in this code block for it to execute properly, and since it is
anonymous, you don’t need execute permission on it, eliminating those answers.
Finally, you don’t have enough information to determine if the user cannot insert
onto the table, because you haven’t been able to compile it yet. (Topic 20.3)
16. D. Nothing; Oracle raises this as an exception automatically.

Explanation Oracle raises an exception automatically when an attempt to select
data into a fetch variable fails. However, to handle that exception, you would use
the no_data_found exception, not rowtype_mismatch, making that choice
incorrect. You also do not need to include any implicit cursor variable for Oracle to
detect this situation, making those other choices including implicit cursor attributes
like %found or %notfound incorrect as well. (Topic 19.5)
17. A. insert into EMPLOYEE values (59694,’HARRIS’, NULL);

Explanation This choice is acceptable because the positional criteria for not
specifying column order are met by the data in the values clause. When you would
like to specify that no data be inserted into a particular column, one method of doing
so is to insert a NULL. Choice B is incorrect because not all columns in the table have
ORACLE
Blind Folio 1:84
values identified. When using positional references to populate column data, there
must be values present for every column in the table. Otherwise, the columns that will
be populated should be named explicitly. Choice C is incorrect because when a column
is named for data insert in the insert into clause, a value must definitely be specified
in the values clause. Choice D is incorrect because using the multiple-row insert
option with a select statement is not appropriate in this situation. (Topic 10.2)
18. D. where C.COW_NAME = ’BESS’ AND C.CARTON_NUM =

C1.CARTON_NUM;
Explanation Two components are required in your where clause—you need
a join clause and something that only pulls records from COW_MILK for BESS.
The right answer is where C.COW_NAME = ’BESS’ AND C.CARTON_NUM =
C1.CARTON_NUM. Another choice is similar to this one, but since it uses the “not
equal” or <> clause for getting information only for BESS, it is not the choice you
want. The other two choices are incomplete, and therefore wrong. (Topic 5.1)
19. D. The inserts contain duplicate data due to the reset sequence.
Explanation The correct answer is that the inserts contain duplicate data due to
the reset sequence. When you drop and re-create the sequence from its original code,
you reset the start value for that sequence. Subsequent insert statements will then
attempt to add rows where the value in the primary key is duplicated information.
There is no information about read-only status in the question, so you should
assume that is not the answer. Dropping a sequence does nothing to a table’s
primary key—there is no relationship between the two. Finally, although it is true
that any cached sequence values that existed when the sequence was dropped are
now unusable, this point has little relevance to the question at hand. (Topic 15.2)
20. B. exists
Explanation Only when using the exists statement must you use a correlated
subquery. Although you can use a subquery with your use of in, you are not
required to do so because you can specify a set of values instead. The between
keyword indicates a range of values and does not allow use of a subquery. The
like keyword is used for wildcard comparisons and also does not allow use of
a subquery. (Topic 7.2)
21. D. abs( )
Explanation All functions except for abs( ) will give you a result of 4093 when
you pass them 4093.505. abs( ) returns the absolute value of the number you pass
into the function. round( ) can do it if you also pass in a second parameter
ORACLE
Blind Folio 1:85
defining the precision to which you want to round the function, while trunc( )
will do it with only 4093.505 as input. floor( ) does it too, because it is the
logical opposite of the ceil( ) function. (Topic 4.2)
22. B. Use accept to capture name value for each run.

Explanation The accept command is the best way to handle the situation.
Although you could use define to assign a value to a variable used throughout the
script, only accept allows you to dynamically enter a value for that variable.
Lexical substitutions identified with the & character will only work for the current
statement, meaning that the same value assigned in one statement will not be used
in the next statement unless you reenter it. (Topic 9.1)
23. B. truncate
Explanation Once a truncate operation is complete, that’s it—the change is
made and saved. This is because truncate is not a DML operation that can be
performed as part of a transaction. If you want to get the data back after truncating,
you have to recover it. For any other operation listed as a choice in this question,
such as insert, update, and delete statements, Oracle allows you to discard
the changes using the rollback command. (Topic 11.5)
24. A. savepoint
Explanation savepoint operations simply act as logical breakpoints in a transaction.
They do not cause Oracle to save or discard data, but merely act as a breakpoint with
which you can perform partial transaction rollbacks later. All other commands, namely
set transaction and commit, indicate the beginning of a new transaction. Creating
a new session with Oracle implicitly begins a transaction as well. (Topic 10.5)
25. B. A from clause

Explanation No SQL statement can survive without a from clause. For this reason,
Oracle provides you with the DUAL table, so that you can perform arithmetic operations
on expressions and not on table data while still satisfying this syntactic construct. Since
this statement already has a select clause, you don’t need to add another. The where
clause is optional, but since the statement already has one, you don’t need to add
another. Finally, your SQL statement does not require an order by clause. (Topic 2.2)
26. D. Values from a cursor cannot be referenced directly.

Explanation The problem lies in the line reading ...VALUES (UPC_CODE_CRSR);.
You cannot reference the cursor directly in this way because a cursor is merely an
address in memory. Instead, you have to reference the values as they are fetched
ORACLE
Blind Folio 1:86
into a variable. The cursor looping mechanism itself is fine and does not need to be
rewritten. The exception handler does not need to be defined for the PL/SQL block,
either. Finally, since this is an anonymous block, there
is no concept of “having permission” to execute it—if you submit it, you can
execute it. (Topic 22.1)
27. C. Code a when others exception handler in PROC_A.

Explanation The others exception handler in PROC_A will prevent exception
propagation to the user level if no exception handler exists in PROC_B. Remember,
Oracle always raises its own exceptions. If that exception is not handled locally, it
is propagated to the next level up, where either there must be an explicit exception
handler or the when others handler. (Topic 18.3)
28. B. DECLARATION, EXECUTION, and EXCEPTION

Explanation The exception you define is a user-defined exception. You must include
support for this type of exception in the declarative, execution, and exception
sections of the PL/SQL block. Support for an Oracle predefined exception needs
only to be included in the exception handler, whereas if you wanted to associate
an internal error with a named exception, you would code support for it in the
declaration and exception section of your code block. (Topic 24.1)
29. A. set transaction

Explanation The set transaction command is used to define the transaction
state to be read-only. rollback and commit statements are used to end the
transaction, while the savepoint command denotes logical breakpoints for the
transaction. (Topic 10.5)
30. C. Bitmap index

Explanation Bitmap indexes work well in situations where the data in the column
is static. In this case, the column contains gender information, which will rarely if
ever change. The number of distinct possible values is limited to only two as well.
Thus, this column is a bad candidate for B-tree indexes of any sort, but perfect for
bitmap indexes. Remember, B-tree indexes work well for columns with high
cardinality or number of distinct values corresponding to the overall number of
entries in the column. (Topic 15.1)
ORACLE
Blind Folio 1:87
31. A. Optional, one-to-many

Explanation This question describes the notation you should use on your logical
data model, and the appropriate answer is optional and one-to-many. This is
because the relationship described between invoices and invoice items is optional,
given that invoices may have no invoice items. The second part of the notation is
the ordinality of the relationship. In this case, one invoice can have many items.
(Topic 12.1)
32. D. ACTIVE
Explanation The decode( ) function acts as a case statement. The first parameter
indicates the column whose values you want decoded. If the value in the column
equals parameter 2, then decode( ) returns parameter 3. If the value in the column
equals parameter 4, decode( ) returns parameter 5, and so on. If the value in the
column doesn’t equal any of the other parameters specified, then decode( )
returns the default value specified as the last parameter. Thus, since the column
value is something not specified for any of the parameters, the returned value
is the default, ACTIVE. (Topic 4.3)
33. B. Tablespaces
Explanation A tablespace, the logical object used for storing database objects like
tables and indexes, maps most directly to a datafile because tablespaces can have one
or many datafiles. Although segments and extents are stored in datafiles, the mapping
is much closer between tablespace and datafile because both are storage containers.
Finally, although both containers are comprised of Oracle blocks, the concept of
Oracle blocks has less meaning at the filesystem level, where the datafile to the OS
will look just like any other file in the host system. (Topic 1.1)
34. B, D, and E. RDBMS defines how to obtain data for you, RDBMS allows
flexibility in changing data relationships, and RDBMS is able to model
relationship other than master/detail.
Explanation A relational database differs from hierarchical databases like IMS in
many ways. First, the RDBMS handles data access methods implicitly within the
engine, shielding users from defining how to access data physically on the machine
(i.e., “open this file, search for this text string,” etc.)—hierarchical systems like IMS
require that you define methods to traverse the master/detail relationships to obtain
information. Both hierarchical databases and RDBMS systems can store data in
master/detail fashion, which eliminates one choice. Also, because hierarchical
ORACLE
Blind Folio 1:88
databases require that you store all data in master/detail format and define the methods
used to access the data, hierarchical databases are less flexible than RDBMS systems
when it comes to changing the way data relates to other data. Finally, RDBMS systems
can model data in relationships other than master/detail. (Topic 1.2)
35. B. Ease of integrating programs written in different languages

Explanation Developers typically don’t use PL/SQL if they need to integrate
programs written in different languages as part of one application. Although later
versions of Oracle allow integration with C using EXTPROC, integrating with other
languages like COBOL, ADA, and FORTRAN generally makes better use of the
Oracle precompilers. However, a developer might choose to use PL/SQL if it is
important to have easy access to SQL, the ability to handle cursors, or the ability to
handle error situations without necessarily having to code an explicit conditional
operation for every type of error encountered. (Topic 1.3)
36. D. where COL_A in (select NUM from TAB_OF_NUMS)

Explanation The where clause in choice D is an excellent example of the definition
of a subquery, which is the example being asked for in the question. Choice A is not
a comparison operation between a column and a set of values, as there is only one
value being compared. Choice B is a comparison of a column to a set of values,
but the set is static and defined at the time the query is issued. Choice C is a range
comparison operation, a variant on choice B, and therefore also wrong. Only
choice D allows Oracle to dynamically generate the list of values to which COL_A
will be compared. (Topic 7.2)
37. D. select NAME from CONTESTANT where COUNTRY in (select

COUNTRY from MEDALS where NUM_GOLD + NUM_SILVER +
NUM_BRONZE > 10)
Explanation The select NAME from CONTESTANT where COUNTRY in (select
COUNTRY from MEDALS where NUM_GOLD + NUM_SILVER + NUM_BRONZE > 10)
query is correct because it contains the subquery that correctly returns a subset of
countries that have contestants who won 10 or more medals of any type. Choice A is
incorrect because it contains a join operation, not a subquery. Choice B is simply a
rewrite of choice A to use a multiple-row subquery, but does not go far enough to
restrict return data. Choice C is a single-row subquery that does essentially the same
thing as choice B. (Topic 7.3)
38. C. Multiple-column subquery, the youngest contestant from all countries

Explanation Since the main query compares against the results of two columns
returned in the subquery, this is a multiple-column subquery that will return the
ORACLE
Blind Folio 1:89
youngest contestant from every country in the table. This multiple-column subquery
is also a multiple-row subquery, but since the defining factor is the fact that two
columns are present, you should focus more on that fact than on the rows being
returned. This fact eliminates choices A and B. The subquery does return multiple
rows, however. You should also be sensitive to the fact that the main query must use
an in clause, not the equals sign (=), making choice D incorrect as well. (Topic 8.1)
39. A. SOO
Explanation The correct answer is SOO because the subquery operation specified
by the in clause ignores NULL values implicitly. Thus, because SOO has no country
defined, that row is not selected as part of the subquery. So, as a result, SOO won’t
show up as having the youngest age for anyone in the results of this query. (Topic 8.2)
40. B. TABLE
Explanation The object being referred to is a table. A table has many columns,
each of which is functionally dependent on the key column. Choice A is incorrect
because a cursor is simply the result of a query, which may or may not have been
against a table, and does not require any kind of storage in a segment. A sequence
is a number generator in Oracle that, again, does not require storage in a segment
other than a dictionary segment, making choice C incorrect. Finally, a view is
similar to a table in that it contains many columns, each of which is functionally
dependent on the key. However, views contain no data needing to be stored in a
segment, so choice D is wrong as well. (Topic 11.1)
41. B. Executable
Explanation The executable section of a PL/SQL block is where the main operation
of the block is written. Thus, any conditional operation such as the one referred to
by the question would be written in the executable section of the PL/SQL block. The
declaration section is used to declare variables, eliminating choice B and choice A.
The exception section is where exception handlers are defined for error
management, eliminating choice D as well. (Topic 17.2)
42. A. The variable is declared incorrectly.

Explanation Variables must be declared in the declaration section of a code block,
making choice A the correct answer. The looping mechanism is set up correctly,
eliminating choice B. The insert statement is correct as well, eliminating choice
C. Finally, the loop needn’t be closed in an exception handler. It can be closed in
the executable section as well. It is simply important that the cursor be closed
somewhere. This fact pattern eliminates choice D. (Topic 17.3)
ORACLE
Blind Folio 1:90
43. D. Using the start command

Explanation The start command in SQL*Plus is designed for executing the contents
of a file. This file may or may not contain any of the other methods listed as choices.
Thus, choice D is the correct answer. The execute command is definitely a way to
execute PL/SQL blocks in SQL*Plus, as is it possible to execute PL/SQL blocks from
within other PL/SQL blocks. These facts eliminate choices A and B. Finally, you can
execute PL/SQL blocks from within select statements so long as the PL/SQL block
being executed is a function. (Topic 17.5)
44. A and D. VAR_1 CONSTANT NUMBER(3) :=96; and VAR_1

NUMBER(3) INITIAL 96;
Explanation Declaring variables with initial values in Oracle PL/SQL is accomplished
either by assigning a value to the variable at the time of declaration with the assignment
operator (:=) or the declare keyword. If you do not use either of these methods,
your initial value declaration will not be correct. Choice A identifies the correct way
to declare a constant, but unfortunately, once declared you cannot assign another
value to it. The initial keyword is used in storage declarations for database
objects, not as part of the PL/SQL command syntax. (Topic 17.4)
45. C. Execution section

Explanation Only the execution section, as denoted by the begin and end
keywords, is required in order for the PL/SQL block to compile. The declaration
section is not required, but if it does appear, it must appear before the execution
section. Thus, choice A is incorrect. The exception section is also not required,
and can appear either at the end of the execution section or at any point within
the execution section. Thus, choice B is incorrect. Finally, the “parameter passing”
section or named PL/SQL block declaration section is not required because
anonymous blocks are permitted in Oracle, thus making choice D incorrect as
well. (Topic 18.1)
46. C. The nested block must be defined after variables used in the main block.
Explanation In PL/SQL, the rule about nested blocks is that the sub-block must be
defined after the variables, cursors, and constants defined for use in the main block.
Thus, choice C is correct. You can have named subprograms in anonymous blocks
in PL/SQL, thus meaning that choice A is incorrect. There is no requirement in PL/SQL
that variables passed to other blocks must have the same name as the parameter
defined in the sub-block. In fact, there are compelling reasons not to do so. Thus,
choice B is incorrect. The default keyword is used as an alternative to the
assignment operator in PL/SQL for giving a declared variable an initial value,
thus making choice D incorrect as well. (Topic 18.3)
ORACLE
Blind Folio 1:91
47. A. Package constants in the local procedure

Explanation You would not use synonyms to identify package constants in the local
procedure. Nor would you need to—these constants would be available in the local
procedure without any scoping difficulties. Choices B, C, and D all identify exactly the
types of situations where synonyms are useful. These situations include reference to
stand-alone procedures and tables in distributed databases, or reference to procedures
and tables in the same database that are owned by different users. (Topic 18.5)
48. D. goto
Explanation Of the control structures given, only the goto command allows you
to jump to a different portion of the PL/SQL lock unconditionally. Even the if true
statement, which always executes the block of code contained in the if-then (and
thus, an if-then-else) structure, is not correct because it does not allow you to
jump to a completely different section of code without running any kind of test. For
this reason, choices A and C are incorrect. Finally, choice B is incorrect because a
loop statement simply iterates through a block of code for as many times as
specified by the looping construct. (Topic 20.1)
49. D. NULL
Explanation Because there is no row in the table where the country is Japan,
the select statement returned no value. Thus, the length of VAR1 will be NULL.
Anything evaluated in conjunction with NULL becomes NULL as well, and thus
VAR2 is assigned when the length of VAR1 plus 55 is executed. Thus, choice D
is the correct answer. (Topic 20.4)
50. A. The label is not followed by an executable statement.

Explanation All labels in PL/SQL statements must be unique within their scope and must
also precede an executable statement. In this case, the label precedes an end loop clause,
which is not an executable statement unto itself. Instead, the end loop clause is part of
the overall iteration construct used to define the loop. Thus, you must have an executable
statement following the label and before the end loop clause. For your information, the
NULL keyword can be used in this context following the label, as NULL by itself followed
by a semicolon actually constitutes an executable statement. (Topic 20.5)
51. B.
TYPE cont_rec IS (
NAME VARCHAR2(20),
AGE NUMBER(5),
ORACLE
Blind Folio 1:92
Explanation The correct answer, shown as choice B, is correct because the question
asks you to define a user-defined record. This simply means you have to first define
the elements of this record using the type declaration command, then define a
variable as that type. Choice A simply has you declaring stand-alone variables for
each of the elements in the record, which will work correctly but is not a user-defined
record. Choice C is a representation of how to define a record using the %rowtype
attribute, which is the most efficient way to define this record but is not a user-defined
record unto itself. Finally, choice D identifies an invalid command syntax. (Topic 21.1)
52. D. CONTESTANT_TABLE IS TABLE OF CONTESTANT%ROWTYPE INDEX

BY BINARY_INTEGER;
Explanation To declare a PL/SQL table of records where each record corresponds to
a particular table in the Oracle database, you can use the %rowtype attribute in the
table declaration itself, thus making choice D correct. Choice A is incorrect because
you cannot define a PL/SQL table of records using native datatypes in the form of a
multidimensional array as the syntax in that choice’s command would suggest. Choice B
is also incorrect because the use of the %type attribute gives only one column datatype,
thus making the PL/SQL table a one-dimensional table. Furthermore, the manner in
which the type declaration is constructed is incorrect in that choice as well. Finally,
choice C is missing the all-important indexing mechanism required in your PL/SQL
table of record, making that choice incorrect as well (Topic 21.4)
53. B and C. Value in MYVAL and CONTESTANT_NAME_TABLE.COUNT

Explanation After exiting the cursor for loop, the value in the MYVAL variable
will represent the number of records in the PL/SQL table, so referencing that variable
is one quick way to get the number of rows in the PL/SQL table. Another is to use
the count attribute, available on every PL/SQL table. While you can develop another
for loop to count the number of rows in the PL/SQL table, as indicated by choice
D, this would require several additional lines of code. This is not as efficient as choices
B or C, so choice D is wrong. Finally, you cannot reference a PL/SQL table using
group functions in SQL the way you can in an Oracle database table. (Topic 21.3)
54. D. PL/SQL table of records

Explanation When dealing with bulk data operations, if you want to use a PL/SQL
variable to hold data from multiple table rows, your best bet is a PL/SQL table of
records. This is because each record can act as a row of data, while the PL/SQL
table indexing mechanism allows you to store each row as an element of an array.
In essence, you store and treat the table data as a two-dimensional array. Choice A
is incorrect because only a single row of data can be stored in a record. Choice B is
incorrect because even though the cursor object can be used to refer to an entire
ORACLE
Blind Folio 1:93
table’s worth of information, the cursor itself is merely an address in memory of an

executed SQL statement, not an actual variable. Choice C is incorrect because a
PL/SQL table is only a one-dimensional array. So, only tables with single columns
could be stored in it. (Topic 21.5)
55. C. Oracle retrieves all records in the CONTESTANT table into the cursor
and locks those rows for update.
Explanation The for update clause allows you to simultaneously collect one
or several rows from a table and place a lock on those rows in the table for change
during the code block. The change will not be made until you issue an update
statement, however, making choice D incorrect. Because Oracle does not try to
update the statement, choice A is incorrect, too. It is important to remember that
the for update clause also locks rows from the original table, or else the cursor
would simply select all the data and choice B would be correct. (Topic 24.2)
56. A. update CONTESTANT SET COUNTRY = ‘USA’ where current

of asy_csr;
Explanation The where current of clause allows you to reference back to the
original table record corresponding to this cursor record without explicitly defining
a where clause to do so in the update statement. Thus, you achieve the same
functionality in choice A as you would get with choice C, but since choice C is
attempting to reference the actual cursor value rather than the variable into which
that cursor record’s values were fetched, choice C is actually invalid and is therefore
incorrect. Choice B is also incorrect because the for update clause only locks the
records for update, it doesn’t actually change any records at all. Finally, because
choice A is the correct answer, choice D is inherently incorrect. (Topic 24.3).
57. D PL/SQL allows cursors to contain the same types of subqueries

permitted in any SQL query.
Explanation Any subquery permitted in a SQL select statement is also fair game for
a cursor. Thus, choices B and C both identify types of subqueries permitted in cursors.
But because both statements are correct, neither choice by itself is the complete answer,
so both choices are wrong. Choice A is also wrong inherently. (Topic 24.4)
58. B. PROC_C, PROC_B, PROC_A

Explanation PL/SQL first looks at the exception section in the current code block
to find a handler for the exception being raised. If none is found, Oracle goes to the
procedure caller’s exception section to find a handler. This process continues either
ORACLE
Blind Folio 1:94
until an exception handler is identified or until the error is returned to the user level.
However, Oracle will not return an error to the user level automatically in any
situation—it will always attempt to identify an exception handler first. (Topic 24.5)
59. D. None; cursor for loops handle cursor opening implicitly.

Explanation The cursor for loops handle, among other things, the opening,
parsing, and executing of named cursors. No other loop construct gives you the
implicit support granted by a cursor for loop, so be sure you know how to construct
these powerful looping mechanisms both for OCP and for your development efforts.
(Topic 22.3)
60. D. All exit conditions for while loops are handled in the exit
when clause.
Explanation There is no need for an exit statement in a while loop, since the
exiting condition is defined in the while statement, eliminating choice A. Choice B
is also wrong because you don’t specifically need to use a counter in a while loop
the way you do in a for loop. Finally, choice C is incorrect because even though
the exit condition for a while loop evaluates to a Boolean value (for example, exit
when (this_condition_is_true), the mechanism to handle the exit does not
require an explicit IF-THEN statement. (Topic 20.3)

1. A. A developer wants records to be added to an EXIT_STATUS table
regardless of whether the transaction completed successfully or not.
Explanation Autonomous transactions allow for procedures to execute their own
transaction within the scope of their procedure, without interfering with the transaction
that may have been going on within the calling procedure. This feature would allow
the developer to create a procedure to add records to an EXIT_STATUS table and
commit those records without disrupting the rollback of the transaction that would
need to occur in the calling procedure if that transaction failed. Normal transaction
processing in Oracle allows two users to make changes to different rows in different
or same tables at the same time. Thus, choices B and C are eliminated. No user can
make a change to the same row at the same time as another user. Table locks in
Oracle prevent it. Thus, choice D is also eliminated. (Topic 19.4)
ORACLE
Blind Folio 1:95
2. B. cube
Explanation The cube keyword included in a group by clause of SQL statements
in Oracle8i allows you to perform N-dimensional cross-tabulations within the Oracle
database, returning the result set directly to the client with ease. This keyword is
useful in queries within data warehouses and makes choice B correct. Choice C is
incorrect because even though the rollup keyword was also added to SQL queries
in Oracle8i, this keyword supports subtotal and grand total calculations of grouped
data. Although the having expression is also available in group operations, choice
A is incorrect because you needn’t define a having clause in order to use either
cube or rollup. Finally, choice D is incorrect because the trim( ) function
combines the abilities of ltrim( ) and rtrim( ). (Topic 6.1)
3. B. group by
Explanation The group by clause is added by Oracle8i within the scope of
summary management implicitly. This allows for query rewrite, and thus enhanced
performance in the data warehouse. The order by clause is not implicitly added,
thus making choice A incorrect. And, though cube and rollup may be used to
add additional information to the output of the grouped query, this information is
not added implicitly by summary management. (Topic 6.1)
4. D. Level hierarchy
Explanation When creating a dimension, two aspects must be defined. First, the
various levels must be defined for the dimension. Second, the hierarchy between
those levels must be defined. cube and rollup operations do not need to be
defined, so choices A and B are incorrect. Also, summary management is not
defined as part of the dimension definition, so choice C is incorrect. (Topic 15.1)
5. D. Nested table
Explanation Oracle recommends that, if individual items in the collection object
must be accessed, you use TABLE; otherwise, use VARRAY. Since you want to access
individual items in the collection object, choice D is correct and choice C is wrong.
Choice A is wrong because although you’ll want to use scalar types in the definition
of the nested table type, the overall type will be a nested table, not scalar. Finally,
choice B is wrong because even though the table type will be user defined, there are
more specific choices available that more accurately reflect the answer. (Topic 21.5)
ORACLE
Blind Folio 1:96
6. B. Reverse-key indexes
Explanation A reverse-key index is one where the contents of the indexed column
are reversed. This gives a higher amount of lead-in selectivity than a straight B-tree
index would, because the cardinality of the root node in the B-tree would be low.
This is based on the fact that most records would begin with 1 (recall the question
content if you don’t understand why), whereas the reverse of that key would have
greater cardinality. Be careful of choice A, because although cardinality is high,
choice B gives a better option for performance. Choice C is incorrect because bitmap
indexes are designed for low-cardinality records like status or gender. Finally, choice
D indicates an index type that wouldn’t suit this situation. (Topic 15.1)
7. A. VARCHAR2
Explanation Since the text blocks are within the size limits imposed in Oracle8i
for the VARCHAR2 datatype, it is best to use the scalar type rather than a LOB for
simplicity sake. If the block was larger than 4,000 bytes, you would most likely use
a CLOB, but since the size requirement is less than 4,000 bytes, choice C is incorrect.
You would use a BLOB to store binary large objects, making choice B incorrect.
Finally, the text block is not stored as an external file (you would not use the BFILE
type), making choice D incorrect. (Topic 11.3)
8. B and D. execute on procedure foobar( ) and select on table

FOO
Explanation Since the procedure contains the line AUTHID CURRENT_USER,
this procedure will use invoker’s rights execution. Thus, FLUFFY not only needs
permission to run the procedure, but the user needs permission to perform all actions
specified within the procedure, as well. Since no procedure is being created, choice A
is incorrect. Since no table is being updated, choice C is incorrect. (Topic 19.4)
9. B. Any uncommitted changes made to table TAB_2 by my_tran( ) will

not be seen by upd_tran( ).
Explanation Any uncommitted transaction in a procedure that calls a second
procedure defined to use autonomous transactions will not be seen by that second
procedure. Since the transaction in upd_tran( ) is autonomous, however, it will
not commit changes made in my_tran( ), eliminating choice A. The upd_tran( )
procedure will also encounter a deadlock if my_tran( ) already has a lock on
TAB_1, eliminating choice C. Finally, this procedure does not define invoker’s rights
execution, eliminating choice D. (Topic 19.4)
ORACLE
Blind Folio 1:97
10. B. There is no value associated with the constant.

Explanation A value must be associated with a constant in the declaration section.
If no value is given for the constant, an error will result. (Topic 17.3)
11. C. Control will pass to the PL/SQL block caller’s exception handler.
Explanation If the exception raised is not handled locally, PL/SQL will attempt to
handle it at the level of the process that called the PL/SQL block. If the exception is
not handled there, PL/SQL will attempt to keep finding an exception handler that
will resolve the exception. If none is found, the error will be returned to the user.
(Topic 24.5)
12. A. Implicit cursors are used for SQL statements that are not named.
Explanation Implicit cursors are used for all SQL statements except for those
statements that are named. They are never incorporated into cursor for loops,
nor is much care given to using them more or less, which eliminates choices B and
C. They are definitely a feature of Oracle, eliminating choice D. (Topic 22.1)
13. D. Requires exit condition to be defined

Explanation A cursor for loop handles just about every feature of cursor
processing automatically, including exit conditions. (Topic 22.3)
14. A. Use EMPLOYEE.LASTNAME%type.

Explanation The only option in this question that allows the developer to use referential
type declarations for columns is choice A. Choice B uses the %rowtype referential
datatype, which defines a record variable and is not what the developer is after.
(Topic 19.2)
15. B. A numeric value representing the number of rows updated

Explanation %rowtype returns the numeric value representing the number of
rows that were manipulated by the SQL statement. (Topic 21.2)
16. A, C, D. %found, %notfound, %rowcount

Explanation These three are the only choices that are valid cursor attributes. The
%too_many_rows attribute does not exist in PL/SQL. The %rowtype is a keyword
that can be used to declare a record variable that can hold all column values from a
particular table. (Topic 19.5)
ORACLE
Blind Folio 1:98
17. D. exit
Explanation Without an exit statement, a simple loop will not stop. Though the
loop and end loop keywords are needed to define the loop, you should assume
these are in place and you are only trying to figure out how to end the loop. The
if-then syntax might be used to determine a test condition for when the loop
execution should terminate, but it is not required in and of itself to end the loop
process execution. (Topic 20.3)
18. B. others
Explanation There is no others exception. The others exception handler handles
all exceptions that may be raised in a PL/SQL block that do not have exception
handlers explicitly defined for them. All other choices identify Oracle predefined
exceptions that are all caught by the others keyword when used in an exception
handler. If there is no specific handler for another named exception, the others
exception handler will handle that exception. (Topic 24.4)
19. C. into action_record

Explanation The into clause is not permitted in cursors, nor is it required. Your
fetch operation will obtain the value in the current cursor record from the cursor.
(Topic 22.1)
20. D. None; cursor for loops handle cursor opening implicitly.

Explanation The cursor for loops handle, among other things, the opening,
parsing, and executing of named cursors. (Topic 22.3)
21. D. All exit conditions for while loops are handled in the exit when
clause.
Explanation There is no need for an exit statement in a while loop, since the exiting
condition is defined in the while statement, eliminating choice A. Choice B is also
wrong because you don’t specifically need to use a counter in a while loop the way
you do in a for loop. Finally, choice C is incorrect because even though the exit
condition for a while loop evaluates to a Boolean value (for example, exit when
(this_condition_is_true), the mechanism to handle the exit does not require
an explicit if-then statement. (Topic 20.3)
22. C. L
Explanation The fourth letter in the word “processing” is C, which means that the
portion of the if-then statement where VAR1 = C will be used to determine the
ORACLE
Blind Folio 1:99
value for VAR2 (L in this case). Thus, VAR2 is set to ‘L’, and then written to the
database with the insert statement at the end. Be careful not to waste time on
reviewing all the intricacies of the PL/SQL block provided. (Topic 20.2)
23. D. VAR1 := VAR2 + 3049;

Explanation The main problem with this block of PL/SQL code has to do with the
VAR1 := VAR2 + 3049 statement. This is because VAR1 cannot be assigned a value
in this code block because the variable is defined as a constant. VAR2 NUMBER := 0;
is a proper variable declaration. The into VAR2 clause is appropriate in a PL/SQL
fetch statement. Finally, the where NAME = ’SMITH’; clause is well constructed.
All other lines of code in the block not identified as choices are syntactically and
semantically correct. (Topic 17.2)
24. B. select GENDER into VAR1 from EMP

Explanation The assignment of a value to an “in” variable defined in function creation
is not permitted, thus making choice B correct. There is a definition of return value
datatype in this code block, making the function declaration line an incorrect
answer. Although it may seem that the if-then statement in the third line of the
code block is incorrect because you are comparing a VARCHAR2 variable to the
number “6”, in reality Oracle handles this situation just fine because there is an
implicit type conversion occurring in the background. Finally, the where clauses
of the fetch statement are all constructed correctly. (Topic 20.2)
25. B. The index will be dropped.

Explanation Like automatically generated indexes associated with a table’s primary
key, the indexes created manually on a table to improve performance will be dropped
if the table is dropped. Choices A, C, and D are therefore invalid. (Topic 15.3)
26. C. Columns with high cardinality are handled well by B-tree indexes.
Explanation Columns with low cardinality are the bane of B-tree indexes,
eliminating choice A. Furthermore, bitmap indexes are primarily used for performance
gains on columns with low cardinality, eliminating choice B. The correct answer is C.
(Topic 15.1)
27. D. Drop and re-create the view with references to select more columns.
Explanation Choice A is incorrect because adding columns to the underlying table
will not add columns to the view, but will likely invalidate the view. Choice B is
incorrect because the alter view statement simply recompiles an existing view
ORACLE
Blind Folio 1:100
definition, whereas the real solution here is to change the existing view definition
by dropping and re-creating the view. Choice C is incorrect because a correlated
subquery will likely worsen performance, and underscores the real problem—a
column must be added to the view. (Topic 13.2)
28. C. maxvalue
Explanation The maxvalue option is a valid option for sequence creation.
Choices A and B are both part of the create user statement, while choice D is a
part of a constraint declaration in an alter table or create table statement.
(Topic 15.2)
29. F. This statement contains no errors.

Explanation Even though the reference to with check option is inappropriate,
considering that inserts into complex views are not possible, the statement will not actually
produce an error when compiled. Therefore, there are no errors in the view. This is not
something that can be learned. It requires hands-on experience with Oracle. (Topic 15.3)
30. A. references
Explanation The references privilege gives the user the ability to refer back
to your table in order to link to it via a foreign key from his or her table to yours.
Choice B is incorrect because the index privilege allows the user to create an index
on a table, while choice C is incorrect because the select privilege allows users
to query data in your table. Finally, choice D is incorrect because the insert
privilege is only required for allowing the other user to insert data into your table.
(Topic 16.3)
31. A, B, and C.
Explanation Choice D is the only option not available to managing roles. Roles
cannot be granted to synonyms. (Topic 16.2)
32. C. Is equal to NEXTVAL

Explanation Once NEXTVAL is referenced, the sequence increments the integer
and changes the value of CURRVAL to be equal to NEXTVAL. (Topic 15.2)
33. D. with check option

Explanation The appropriate clause is with check option. You can add this clause
to a create view statement so that the view will not allow you to add rows to the
ORACLE
Blind Folio 1:101
underlying table that cannot then be selected in the view. The with {admin|grant}
option clauses are used to assign administrative ability to users along with granting
them a privilege. The with security option is a work of fiction—it does not exist
in Oracle. (Topic 16.4)
34. C. The statement fails due to PRIMARY KEY constraint.

Explanation It should be obvious that the statement fails—the real question here is
why. The reason is because of the primary-key constraint on UPC_CODE. As soon
as you try to add a duplicate record, the table will reject the addition. Although the
view has with check option specified, this is not the reason the addition fails. It
would be the reason an insert fails if you attempt to add a record for a day other
than today, however. (Topic 13.4)
35. A. drop view

Explanation When a table is dropped, Oracle eliminates all related database
objects, such as triggers, constraints, and indexes—except for views. Views are
actually considered separate objects, and although the view will not function
properly after you drop the underlying table, Oracle will keep the view around
after the table is dropped. (Topic 13.5)
36. C. The select will receive NO ROWS SELECTED.

Explanation Although the query will succeed (translation—you won’t receive an
error), you must beware of the distracter in choice B. In reality, choice C is the better
answer because it more accurately identifies what really will occur when you issue
this statement. This view will behave as any select statement would when you list
criteria in the where clause that no data satisfies, by returning NO ROWS SELECTED.
This is not an error condition, but you wouldn’t call it a successful search for data,
either, making both those choices incorrect. Finally, select statements never add
data to a table. (Topic 13.3)
37. B and D. UNIQUE constraints and primary keys

Explanation Every constraint that enforces uniqueness creates an index to assist
in the process. The two integrity constraints that enforce uniqueness are UNIQUE
constraints and primary keys. (Topic 12.2)
38. C. ALL_IND_COLUMNS
Explanation This view is the only one listed that provides column positions in an
index. Since primary keys create an index, the index created by the primary key
ORACLE
Blind Folio 1:102
will be listed with all the other indexed data. Choice A is incorrect because no
view exists in Oracle called ALL_PRIMARY_KEYS. Choice B is incorrect because
although USER_CONSTRAINTS lists information about the constraints in a database,
but it does not contain information about the index created by the primary key.
Choice D is incorrect because ALL_TABLES contains no information related to the
position of a column in an index. (Topic 14.2)
39. C. A table named ANIMALS will be created in the ANJU schema with the
Explanation This question requires you to look carefully at the create table
statement in the question and to know some things about table creation. First, a table
is always created in the schema of the user who created it. Second, since the create
table as select clause was used, choices B and D are both incorrect because they
identify the table being created as something other than ANIMALS, among other things.
Choice A identifies the schema into which the ANIMALS table will be created as
MASTER, which is incorrect for the reasons just stated. (Topic 11.2)
40. A. insert into EMPLOYEE values (59694,’HARRIS’, NULL);

Explanation This choice is acceptable because the positional criteria for not
specifying column order are met by the data in the values clause. When you
would like to specify that no data be inserted into a particular column, one method
of doing so is to insert a NULL. Choice B is incorrect because not all columns in the
table have values identified. When using positional references to populate column
data, there must be values present for every column in the table. Otherwise, the
columns that will be populated should be named explicitly. Choice C is incorrect
because when a column is named for data insert in the insert into clause, a value
must definitely be specified in the values clause. Choice D is incorrect because
using the multiple row insert option with a select statement is not appropriate
in this situation. (Topic 10.2)
41. A. Two tables in the database are named VOUCHER and

VOUCHER_ITEM, respectively.
Explanation This choice implies the use of a naming convention similar to the
one discussed. Although there is no guarantee that these two tables are related, the
possibility is strongest in this case. Choice B implies the same naming convention,
and since the two tables’ names are dissimilar, there is little likelihood that the two
tables are related in any way. Choice C is incorrect because the date a table is created
has absolutely no bearing on what function the table serves in the database. Choice
D is incorrect because two tables cannot be related if there are no common
columns between them. (Topic 12.1)
ORACLE
Blind Folio 1:103
42. A, B, and D. CHAR, VARCHAR2, and NUMBER

Explanation BOOLEAN is the only invalid datatype in this listing. Although
BOOLEAN is a valid datatype in PL/SQL, it is not a datatype available on the
Oracle database, meaning that you cannot create a column in a table that uses
the BOOLEAN datatype. (Topic 11.3)
43. D. The delete statement will remove all records from the table.
Explanation There is only one effect produced by leaving off the where clause
from any statement that allows one—the requested operation is performed on all
records in the table. (Topic 10.4)
44. C. GOOD_NAME varchar2(20) check(GOOD_NAME in (select

NAME from AVAIL_GOODS)),
Explanation A CHECK constraint cannot contain a reference to another table, nor
can it reference a virtual column, such as ROWID or SYSDATE. The other lines of
the create table statement contain correct syntax. (Topic 11.2)
45. A. Locks
Explanation Locks are the mechanisms that prevent more than one user at a time
from making changes to the database. All other options refer to the commands that
are issued to mark the beginning, middle, and end of a transaction. (Topic 10.5)
46. A. Use the alter table statement.

Explanation The alter table statement is the only choice offered that allows
the developer to increase the number of columns per table. Choice B is incorrect
because setting a column to all NULL values for all rows does simply that. Choice C
is incorrect because increasing the adjacent column sizes simply increases the sizes
of the columns, and choice D is incorrect because the listed steps outline how to
add a column with a NOT NULL constraint, something not specified by the question.
(Topic 11.4)
47. B. Truncate the table.

Explanation Choices A and C may work, but an upgrade of hardware and software
will cost far more than truncating the table. Choice D is partly correct, as there will
be some change required to the high-water mark, but the change is to reset, not
eliminate entirely, and the method used is to truncate the table. (Topic 11.5)
ORACLE
Blind Folio 1:104
48. A. UNIQUE
Explanation Only UNIQUE and PRIMARY KEY constraints require Oracle to
generate an index that supports or enforces the uniqueness of the column values.
Foreign keys do not require this sort of index. CHECK constraints also do not require
an index. Finally, NOT NULL constraints do not require an index either. (Topic 12.1)
49. B. All values in the referenced column in the parent table must be present
Explanation Referential integrity is from child to parent, not vice versa. The parent
table can have many values that are not present in child records, but the child record
must correspond to something in the parent. Thus, the correct answer is all values in
the referenced column in the parent table must be present in the referencing column
in the child. (Topic 12.1)
50. B. Values must be part of a fixed set defined by create or alter table
Explanation A CHECK constraint may only use fixed expressions defined when you
create or alter the table with the constraint definition. The reserved words like SYSDATE
and USER, or values from a lookup table, are not permitted, making those answers
incorrect. Finally, NULL values in a column are constrained by NOT NULL constraints,
a relatively unsophisticated form of check constraints. (Topic 11.2)
51. B. sqrt( )
Explanation Square root operations are performed on one column value. (Topic 6.3)
52. B. The tables in the join need to have common columns.

Explanation It is possible that a join operation will produce no return data, just as
it is possible for any select statement not to return any data. Choices A, C, and D
represent the spectrum of possibilities for shared values that may or may not be
present in common columns. However, joins themselves are not possible without
two tables having common columns. (Topic 5.1)
53. D. Until the session completes

Explanation A variable defined by the user during a session with SQL*Plus will
remain defined until the session ends or until the user explicitly undefines the
variable. (Topic 9.1)
ORACLE
Blind Folio 1:105
54. B and C. Alter the prompt clause of the accept command and enter a
new prompt in the login.sql file.
Explanation Choice D should be eliminated immediately, leaving the user to select
between A, B, and C. Choice A is incorrect because config.ora is a feature associated
with Oracle’s client/server network communications product. Choice C is correct,
because you can use the set sqlprompt command within your login.sql file.
This is a special file Oracle users can incorporate into their use of Oracle that will
automatically configure aspects of the SQL*Plus session, such as the default text
editor, column and NLS data formats, and other items. (Topic 9.5)
55. A and C.
Explanation Choice A details the use of a runtime variable that can be used to
have the user input appropriate search criteria after the statement has begun processing.
Choice C details the use of a subquery that allows the user to select unknown search
criteria from the database using known methods for obtaining the data. Choice B is
incorrect because the statement simply provides a known search criterion; choice D
is incorrect because it provides no search criteria at all. (Topic 9.1)
56. A. Ampersand
Explanation The ampersand (&) character is used by default to define runtime
variables in SQL*Plus. (Topic 9.1)
57. C. select e.empid, d.head from EMPLOYEE e, dept d where

e.dept# = d.dept# (+);
Explanation Choice C details the outer join operation most appropriate to this user’s
needs. The outer table in this join is the DEPT table, as identified by the (+) marker next
to the DEPT# column in the comparison operation that defines the join. (Topic 5.2)
58. B, C, and D. To exclude certain data groups based on known criteria, to

include certain data groups based on unknown criteria, and to include
certain data groups based on known criteria
Explanation All exclusion or inclusion of grouped rows is handled by the having
clause of a select statement. Choice A is not an appropriate answer because sort
order is given in a select statement by the order by clause. (Topic 6.4)
ORACLE
Blind Folio 1:106
59. B. Produced as a result of a join select statement with no where clause

Explanation A Cartesian product is the result dataset from a select statement where
all data from both tables is returned. Some potential causes of a Cartesian product
include not specifying a where clause for the join select statement. (Topic 5.1)
60. D. Issuing the set define command

Explanation Choice A is incorrect because a change to the initsid.ora file will
alter the parameters Oracle uses to start the database instance. Use of this feature
will be covered in the next unit. Choice B is incorrect because although the
login.sql file can define many properties in a SQL*Plus session, the character
that denotes runtime variables is not one of them. Choice C is incorrect because the
define command is used to define variables used in a session, not an individual
statement. (Topic 9.2

MCQ Book PDF

Uploaded by

Copyright:

Available Formats

MCQ Book PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MCQ Book PDF

Uploaded by

Copyright:

Available Formats

What are the main topic areas covered in OCP Exam 1?

What are the main topic areas covered in OCP Exam 1?

What are the different types of joins that can be performed in SQL?

What are the different types of joins that can be performed in SQL?

ORACLE

Color profile: Generic CMYK printer Series

4 Oracle8i Certified Professional DBA Practice Exams

CP Exam 1 in the Oracle DBA track covers concepts and practices

OCP Exam 1 Topic Areas

1. Overview of Relational Databases, SQL, and PL/SQL

Chapter 1: OCP Exam 1: Introduction to SQL and PL/SQL 5

5. Displaying Data from Multiple Tables

6 Oracle8i Certified Professional DBA Practice Exams

10. Manipulating Data

Chapter 1: OCP Exam 1: Introduction to SQL and PL/SQL 7

15.3. Create and maintain indexes

8 Oracle8i Certified Professional DBA Practice Exams

20.3. Construct and identify different loop statements

Chapter 1: OCP Exam 1: Introduction to SQL and PL/SQL 9

10 Oracle8i Certified Professional DBA Practice Exams

Chapter 1: OCP Exam 1: Introduction to SQL and PL/SQL 11

12 Oracle8i Certified Professional DBA Practice Exams

Chapter 1: OCP Exam 1: Introduction to SQL and PL/SQL 13

C. insert into BANK_ACCT (ACCTNO, NAME) VALUES (&VAR1,

14 Oracle8i Certified Professional DBA Practice Exams

Chapter 1: OCP Exam 1: Introduction to SQL and PL/SQL 15

16 Oracle8i Certified Professional DBA Practice Exams

C. insert into view

Chapter 1: OCP Exam 1: Introduction to SQL and PL/SQL 17

18 Oracle8i Certified Professional DBA Practice Exams

Chapter 1: OCP Exam 1: Introduction to SQL and PL/SQL 19

A. The statement fails due to with check option clause.

20 Oracle8i Certified Professional DBA Practice Exams

Chapter 1: OCP Exam 1: Introduction to SQL and PL/SQL 21

22 Oracle8i Certified Professional DBA Practice Exams

Chapter 1: OCP Exam 1: Introduction to SQL and PL/SQL 23

D. select A.STUDENT_ID, A.LOCATION, B.LOCATION from

24 Oracle8i Certified Professional DBA Practice Exams

Chapter 1: OCP Exam 1: Introduction to SQL and PL/SQL 25

WHERE EMPID = '40593';

26 Oracle8i Certified Professional DBA Practice Exams

B. select to_char(nvl(sqrt(59483), 0)) from dual;

Chapter 1: OCP Exam 1: Introduction to SQL and PL/SQL 27

28 Oracle8i Certified Professional DBA Practice Exams

Chapter 1: OCP Exam 1: Introduction to SQL and PL/SQL 29

30 Oracle8i Certified Professional DBA Practice Exams

C. select * from EMP where EMPID IN (select EMPID from

Chapter 1: OCP Exam 1: Introduction to SQL and PL/SQL 31

8. You create the following PL/SQL block:

32 Oracle8i Certified Professional DBA Practice Exams

D. if VAR1 = 6 then RETURN (6) else return (8);

Chapter 1: OCP Exam 1: Introduction to SQL and PL/SQL 33

34 Oracle8i Certified Professional DBA Practice Exams

D. A loop exit condition must be defined.

Chapter 1: OCP Exam 1: Introduction to SQL and PL/SQL 35

D. where C.COW_NAME = ’BESS’ AND C.CARTON_NUM =

36 Oracle8i Certified Professional DBA Practice Exams

Chapter 1: OCP Exam 1: Introduction to SQL and PL/SQL 37

38 Oracle8i Certified Professional DBA Practice Exams