Introduction to PL/SQL Lecture 0 – Self Study Akhtar Ali Overview Overview of PL/SQL Development of a coded block Interacting with an Oracle database Controlling PL/SQL process flow Cursor handling Error handling Re-visiting SQL Instructions to Oracle identifying the information you wish to select, insert, delete or update SQL*Plus is Oracle's simple interface for interaction with Oracle DBMS using SQL. Notes on SQL are on Blackboard PL/SQL - introduction Procedural extension allowing for modularity, variable declaration, loops and logical constructs. Allows for advanced error handling Communicates natively with other oracle database objects. Managed centrally within the Oracle database. Other Databases All have procedural facilities SQL is not functionally complete Lacks full facilities of a programming language So top up functionality by embedding SQL in a procedural language PL/SQL techniques are specific to Oracle but procedures and functions can be ported to other systems Why use PL/SQL Manage business rules – through middle layer application logic. Generate code for triggers Generate code for interface Enable database-centric client/server applications Centralised V’s De-centralised Begin : Server End; Common copy of executed code – one copy to maintain Begin Begin Begin : : : End; End; End; Multiple copies of executable code on the decentralised system – multiple copies to maintain leading to increase difficulty in maintaining the system Server Advantages of using PL/SQL to access Oracle PL/SQL is managed centrally within the database Code is managed by the DBA and execution privileges are managed in the same way as with other objects PL/SQL objects are first-class Oracle DB objects Easy to read With modularity features and error handling Centralised control Enables DBA to: Specify rules in one place (as procedure, function, package in PL/SQL) Force user access through the predefined PL/SQL so users cannot write their own procedural code and use this instead. Define for instance security privileges giving users access to table(s) only through a particular procedure Using PL/SQL as a programming language Permits all operations of standard programming languages e.g. Provides loops for controlling iteration Conditions IF-THEN-ELSE-END IF; Jumps GOTO LOOP-EXIT; WHEN-END LOOP; FOR-END LOOP; WHILE-END LOOP Allows extraction of data into variables and its subsequent manipulation Modules in PL/SQL There are 4 types of modules in PL/SQL Procedures – series of statements may or may not return a value Functions – series of statements must return a single value Triggers – series of PL/SQL statements (actions) executing after an event has triggered a condition (ECA) Packages – collection of procedures and function that has 2 parts: a listing and a body. Procedures Creation command Create or replace procedure sample1 as Variable declarations v_num1 constant number := 2.5; v_num2 constant number := 4; v_product number; Body of procedure BEGIN v_product := v_num1 * v_num2; END; Use of Data-Types Number – used to store any number Char(size) & varchar2(size) e.g.: char(10) – used to store alphanumerical text strings, the char data type will pad the value stored to the full length declared. Date – used to store dates Long – used to store large blocks of text up to 2 gigabytes in length (limited operations) More data-types Long raw – stores large blocks of data stored in binary format Raw – stores smaller blocks of data in binary formal Rowid – used to store the special format of rowids in the database Non-database Data Types DEC, DECIMAL, REAL, INTEGER, INT – these are numerical data types that are a subset of number. Binary_integer – binary format for number type but can not be stored in database unless converted first. Character – same as char Boolean – true/false value Table/record – tables can be used to store the equivalent of an array while records store the variables with composite data types. Using SQL in procedures Select values into PL/SQL variables using INTO Record.element notation will address components of tuples (dot notation) %rowtype allows full rows to be selected into one variable V_employee employee%rowtype Empid empname addr1 addr2 addr3 postcode grade salary Example (Anonymous Block of Code) Declare v_employee employee%rowtype; Begin select * Selects entire row of data into 1 variable called v_employee into v_employee from employee where empid = 65284; update employee set salary = v_employee.salary + 1000 where empid = v_employee.empid; End; Is updating the value of salary based on selected element of a variable Cursor overview Very powerful in PL/SQL modules Allows more than one set of data to be retrieved and accessed at the same time in loops Sets are created by executing SQL statements embedded in the PL/SQL code Cursor attributes - %notfound, %rowcount, %found & %isopen Error handling Prevents database locking Ensures that errors are presented to the user in a sensible format Makes code robust Essential when using PL/SQL as formal programming language or interfacing with Oracle applications.