Oracle 1Z0-183 key terms

The basic unit of PL/SQL code, made up of an optional declaration section, a mandatory executable section, and an optional exception-handling section, ending with END.

A PL/SQL block without a name that is not stored in the database; it is compiled and run once, usually submitted from a client tool.

The optional part of a PL/SQL block, starting with DECLARE, where variables, constants, cursors, types, and local subprograms are defined before executable statements run.

A variable that holds one value of a single data type, such as NUMBER, VARCHAR2, DATE, or BOOLEAN, rather than a collection or a record.

An attribute that gives a variable the same data type as a specified table column or another variable, so the declaration stays correct when the referenced type changes.

A variable created in the host environment, such as SQL*Plus, that PL/SQL references with a colon prefix and that can pass values into and out of a block.

The internal cursor Oracle opens automatically for every SQL data-manipulation statement run from PL/SQL, whose status is exposed through attributes such as SQL%ROWCOUNT and SQL%FOUND.

An implicit cursor attribute that returns the number of rows affected by the most recently run SQL statement, or the number fetched so far by a cursor.

A clause added to INSERT, UPDATE, or DELETE that returns affected column values into PL/SQL variables, avoiding a separate query to retrieve them.

A clause that fetches multiple rows from a query into one or more collections in a single operation, reducing context switches between the PL/SQL and SQL engines.

A PL/SQL statement that sends one DML operation for all elements of a collection in a single SQL call, improving performance compared with DML inside a loop.

A conditional control structure that runs one of several statement groups depending on which Boolean condition is true first, with an optional ELSE branch when none match.

A control structure that returns one value chosen from one or more WHEN branches, evaluated either by comparing a selector or by testing Boolean conditions.

A loop that tests a Boolean condition before each iteration and keeps running its body while the condition remains true; it may run zero times.

A loop that iterates over a fixed range of integers or over rows returned by a cursor, with the loop index or record implicitly declared and scoped to the loop.

A statement that ends the current loop iteration immediately and transfers control to the next iteration, optionally only when a specified WHEN condition is true.

A composite data type that groups related fields, each with its own data type, under one name so they can be handled as a single variable.

An attribute that declares a record variable whose fields match the columns of a table, view, or explicit cursor, allowing a row to be fetched into one structure.

A collection type whose elements are accessed by a unique index key, which may be an integer or a string, and which is not stored in the database schema.

A variable-size array collection type with a fixed maximum number of elements and sequential integer subscripts, which can also be stored as a database column.

A named work area declared in PL/SQL for a query, which the developer opens, fetches rows from, and closes, with attributes reporting its current state.

A loop that implicitly opens a cursor, fetches each row into a record, runs the loop body, and closes the cursor when no rows remain.

A property such as %FOUND, %NOTFOUND, %ISOPEN, or %ROWCOUNT that reports the state or row count of an explicit or implicit cursor.

A named error condition that Oracle raises automatically for common runtime problems, such as NO_DATA_FOUND or ZERO_DIVIDE, and that developers can handle by name.

An exception declared by the developer in a PL/SQL block or package, raised explicitly with RAISE, and handled in an exception section like a predefined error.

A built-in procedure that stops execution and returns a user-defined error number and message to the caller, using numbers from minus 20000 through minus 20999.

The process by which an unhandled exception moves outward from the block where it occurred to enclosing blocks or the calling environment until a handler catches it.

A named PL/SQL subprogram stored in the database that performs an action, can accept parameters, and is invoked as a standalone statement rather than inside an expression.

A named PL/SQL subprogram stored in the database that computes and returns a value, so it can be called wherever an expression of its return type is allowed.

The public interface of a PL/SQL package, declaring the subprograms, types, variables, and constants visible to callers; it must be created before the package body.

The implementation part of a PL/SQL package, containing the code for public subprograms plus private declarations that are hidden from callers outside the package.

An Oracle-supplied package used to write text to an output buffer that a client tool can display, commonly for debugging and simple reporting from PL/SQL.

An Oracle-supplied package that lets PL/SQL read from and write to operating-system text files on the database server, subject to directory object permissions.

A database trigger that fires automatically in response to INSERT, UPDATE, or DELETE statements on a table or view, running before or after each row or statement.

A trigger defined on a view that runs in place of the triggering DML statement, commonly used to make a non-updatable view support inserts, updates, and deletes.

Pseudo-records available in row-level triggers that expose the new column values being written and the old values being replaced or deleted.

A table that a row-level trigger queries or modifies while that same table is being changed by the triggering statement, which can cause an Oracle runtime error.

A relationship in which one stored PL/SQL object relies on another, such as a procedure referencing a table or a package body calling a function, and which affects recompilation.

The state of a stored PL/SQL object whose dependent objects have changed, meaning it must be recompiled successfully before it can be executed again.

A compiler feature that includes or excludes PL/SQL source text based on selection directives such as $IF, $THEN, and $ELSE, often for version-specific or debugging code.