[vertical list of authors]
©    Copyright   ,.
[cover art/text goes here]




Copyright
   
Contents


Copyright
Apache Software FoundationDerby Reference ManualApache Derby
Copyright
Copyright 1997, 2006 The Apache Software Foundation or its licensors, as applicable.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.
About this guide
For general information about the Derby documentation, such as a complete list of books, conventions, and further reading, see Getting Started with Derby .
Purpose of this document
This book, the Derby Reference Manual , provides reference information about Derby. It covers Derby's SQL language, the Derby implementation of JDBC, Derby system catalogs, Derby error messages, Derby properties, and SQL keywords.
Audience
This book is a reference for Derby users, typically application developers. Derby users who are not familiar with the SQL standard or the Java programming language will benefit from consulting books on those topics.
Derby users who want a how-to approach to working with Derby or an introduction to Derby concepts should read the Derby Developer's Guide .
How this guide is organized
This guide includes the following sections:
 
SQL language reference
Reference information about Derby's SQL language, including manual pages for statements, functions, and other syntax elements.
 
SQL reserved words
SQL keywords beyond the standard SQL-92 keywords.
 
Derby support for SQL-92 features
A list of SQL-92 features that Derby does and does not support.
 
Derby System Tables
Reference information about the Derby system catalogs.
 
Derby exception messages and SQL states
Information about Derby exception messages.
 
JDBC Reference
Information about Derby's implementation of the JDBC interface include support for JDBC 2.0 features.
 
Setting attributes for the database connection URL
Information about the supported attributes to Derby's JDBC database connection URL.
 
J2EE Compliance: Java Transaction API and javax.sql Extensions
Information about the supported attributes to Derby's support for the Java Transaction API.
 
Derby API
Notes about proprietary APIs for Derby.
SQL language reference
Derby implements an SQL-92 core subset, as well as some SQL-99 features.
This section provides an overview of the current SQL language by describing the statements, built-in functions, data types, expressions, and special characters it contains.
Capitalization and special characters
Using the classes and methods of JDBC, you submit SQL statements to Derby as strings. The character set permitted for strings containing SQL statements is Unicode. Within these strings, the following rules apply:
 
Double quotation marks delimit special identifiers referred to in SQL-92 as delimited identifiers.
 
Single quotation marks delimit character strings.
 
Within a character string, to represent a single quotation mark or apostrophe, use two single quotation marks. (In other words, a single quotation mark is the escape character for a single quotation mark.)
A double quotation mark does not need an escape character. To represent a double quotation mark, simply use a double quotation mark. However, note that in a Java program, a double quotation mark requires the backslash escape character.
Example:
-- a single quotation mark is the escape character -- for a single quotation mark VALUES 'Joe''s umbrella' -- in ij, you don't need to escape the double quotation marks VALUES 'He said, "hello!"' n = stmt.executeUpdate( "UPDATE aTable setStringcol = 'He said, \"hello!\"'");
 
SQL keywords are case-insensitive. For example, you can type the keyword SELECT as SELECT, Select, select, or sELECT.
 
SQL-92-style identifiers are case-insensitive (see SQL92Identifier ), unless they are delimited.
 
Java-style identifiers are always case-sensitive.
 
* is a wildcard within a SelectExpression . See The * wildcard . It can also be the multiplication operator. In all other cases, it is a syntactical metasymbol that flags items you can repeat 0 or more times.
 
% and _ are character wildcards when used within character strings following a LIKE operator (except when escaped with an escape character). See Boolean expression .
 
Two dashes (--) and a newline character delimit a comment, as per the SQL-92 standard. The two dashes start the comment and the newline character ends the comment.
SQL identifiers
An identifier is the representation within the language of items created by the user, as opposed to language keywords or commands. Some identifiers stand for dictionary objects, which are the objects you create- such as tables, views, indexes, columns, and constraints- that are stored in a database. They are called dictionary objects because Derby stores information about them in the system tables, sometimes known as a data dictionary. SQL also defines ways to alias these objects within certain statements.
Each kind of identifier must conform to a different set of rules. Identifiers representing dictionary objects must conform to SQL-92 identifier rules and are thus called SQL92Identifier s.
Rules for SQL92 identifiers
Ordinary identifiers are identifiers not surrounded by double quotation marks. Delimited identifiers are identifiers surrounded by double quotation marks.
An ordinary identifier must begin with a letter and contain only letters, underscore characters (_), and digits. The permitted letters and digits include all Unicode letters and digits, but Derby does not attempt to ensure that the characters in identifiers are valid in the database's locale.
A delimited identifier can contain any characters within the double quotation marks. The enclosing double quotation marks are not part of the identifier; they serve only to mark its beginning and end. Spaces at the end of a delimited identifier are insignificant (truncated). Derby translates two consecutive double quotation marks within a delimited identifier as one double quotation mark-that is, the "translated" double quotation mark becomes a character in the delimited identifier.
Periods within delimited identifiers are not separators but are part of the identifier (the name of the dictionary object being represented).
So, in the following example:
"A.B"
is a dictionary object, while
"A"."B"
is a dictionary object qualified by another dictionary object (such as a column named "B" within the table "A").
SQL92Identifier
An SQL92Identifier is a dictionary object identifier that conforms to the rules of SQL-92. SQL-92 states that identifiers for dictionary objects are limited to 128 characters and are case-insensitive (unless delimited by double quotes), because they are automatically translated into uppercase by the system. You cannot use reserved words as identifiers for dictionary objects unless they are delimited. If you attempt to use a name longer than 128 characters, SQLException X0X11 is raised.
Derby defines keywords beyond those specified by the SQL-92 standard (see SQL reserved words ).
Example
-- the view name is stored in the -- system catalogs as ANIDENTIFIER CREATE VIEW AnIdentifier (RECEIVED) AS VALUES 1 -- the view name is stored in the system -- catalogs with case intact CREATE VIEW "ACaseSensitiveIdentifier" (RECEIVED) AS VALUES 1
This section describes the rules for using SQL92Identifiers to represent the following dictionary objects.
Qualifying dictionary objects
Since some dictionary objects can be contained within other objects, you can qualify those dictionary object names. Each component is separated from the next by a period. An SQL92Identifier is "dot-separated." You qualify a dictionary object name in order to avoid ambiguity.
column-Name
In many places in the SQL syntax, you can represent the name of a column by qualifying it with a table-Name or correlation-Name.
In some situations, you cannot qualify a column-Name with a table-Name or a correlation-Name, but must use a Simple-column-Name instead. Those situations are:
 
creating a table ( CREATE TABLE statement )
 
specifying updatable columns in a cursor
 
in a column's correlation name in a SELECT expression (see SelectExpression )
 
in a column's correlation name in a TableExpression (see TableExpression )
You cannot use correlation-Names for updatable columns; using correlation-Names in this way will cause an SQL exception. For example:
SELECT c11 AS col1, c12 AS col2, c13 FROM t1 FOR UPDATE of c11,c13
In this example, the correlation-Name col1 FOR c11 is not permitted because c11 is listed in the FOR UPDATE list of columns. You can use the correlation-Name FOR c12 because it is not in the FOR UPDATE list.
Example
-- C.Country is a column-Name qualified with a -- correlation-Name . SELECT C.Country FROM APP.Countries C
correlation-Name
A correlation-Name is given to a table expression in a FROM clause as a new name or alias for that table. You do not qualify a correlation-Name with a schema-Name.
You cannot use correlation-Names for updatable columns; using correlation-Names in this way will cause an SQL exception. For example:
SELECT c11 AS col1, c12 AS col2, c13 FROM t1 FOR UPDATE of c11,c13
In this example, the correlation-Name col1 FOR c11 is not permitted because c11 is listed in the FOR UPDATE list of columns. You can use the correlation-Name FOR c12 because it is not in the FOR UPDATE list.
Example
-- C is a correlation-Name SELECT C.NAME FROM SAMP.STAFF C
new-table-Name
A new-table-Name represents a renamed table. You cannot qualify a new-table-Name with a schema-Name.
Example
-- FlightBooks is a new-table-Name that does not include a schema-Name RENAME TABLE FLIGHTAVAILABILITY TO FLIGHTAVAILABLE
schemaName
A schemaName represents a schema. Schemas contain other dictionary objects, such as tables and indexes. Schemas provide a way to name a subset of tables and other dictionary objects within a database.
You can explicitly create or drop a schema. The default user schema is the APP schema (if no user name is specified at connection time). You cannot create objects in schemas starting with SYS.
Thus, you can qualify references to tables with the schema name. When a schemaName is not specified, the default schema name is implicitly inserted. System tables are placed in the SYS schema. You must qualify all references to system tables with the SYS schema identifier. For more information about system tables, see Derby System Tables .
A schema is hierarchically the highest level of dictionary object, so you cannot qualify a schemaName.
Syntax
SQL92Identifier
Example
-- SAMP.EMPLOYEE is a table-Name qualified by a schemaName SELECT COUNT(*) FROM SAMP.EMPLOYEE -- You must qualify system catalog names with their schema, SYS SELECT COUNT(*) FROM SYS.SysColumns
Simple-column-Name
A Simple-column-Name is used to represent a column when it cannot be qualified by a table-Name or correlation-Name. This is the case when the qualification is fixed, as it is in a column definition within a CREATE TABLE statement.
Example
-- country is a Simple-column-Name CREATE TABLE CONTINENT (COUNTRY VARCHAR(26) NOT NULL PRIMARY KEY, COUNTRY_ISO_CODE CHAR(2), REGION VARCHAR(26))
synonym-Name
A synonym-Name represents a synonym for a table or a view. You can qualify a synonym-Name with a schema-Name.
table-Name
A table-Name represents a table. You can qualify a table-Name with a schemaName.
Example
-- SAMP.PROJECT is a table-Name that includes a schemaName SELECT COUNT(*) FROM SAMP.PROJECT
view-Name
A view-Name represents a table or a view. You can qualify a view-Name with a schema-Name.
Example
-- This is a View qualified by a schema-Name SELECT COUNT(*) FROM SAMP.EMP_RESUME
index-Name
An index-Name represents an index. Indexes live in schemas, so you can qualify their names with schema-Names. Indexes on system tables are in the SYS schema.
Example
DROP INDEX APP.ORIGINDEX; -- OrigIndex is an index-Name without a schema-Name CREATE INDEX ORIGINDEX ON FLIGHTS (ORIG_AIRPORT)
constraint-Name
You cannot qualify constraint-names.
Example
-- country_fk2 is a constraint name CREATE TABLE DETAILED_MAPS (COUNTRY_ISO_CODE CHAR(2) CONSTRAINT country_fk2 REFERENCES COUNTRIES)
cursor-Name
A cursor-Name refers to a cursor. No SQL language command exists to assign a name to a cursor. Instead, you use the JDBC API to assign names to cursors or to retrieve system-generated names. For more information, see the Derby Developer's Guide . If you assign a name to a cursor, you can refer to that name from within SQL statements.
You cannot qualify a cursor-Name.
Example
stmt.executeUpdate("UPDATE SAMP.STAFF SET COMM = " + "COMM + 20 " + "WHERE CURRENT OF " + ResultSet.getCursorName());
TriggerName
A TriggerName refers to a trigger created by a user.
Example
DROP TRIGGER TRIG1
AuthorizationIdentifier
User names within the Derby system are known as authorization identifiers. The authorization identifier represents the name of the user, if one has been provided in the connection request. The default schema for a user is equal to its authorization identifier. User names can be case-sensitive within the authentication system, but they are always case-insensitive within Derby's authorization system unless they are delimited. For more information, see the Derby Developer's Guide .
Example
CALL SYSCS_UTIL.SYSCS_SET_DATABASE_PROPERTY( 'derby.database.fullAccessUsers', 'Amber,FRED')
Statements
This section provides manual pages for both high-level language constructs and parts thereof. For example, the CREATE INDEX statement is a high-level statement that you can execute directly via the JDBC interface. This section also includes clauses, which are not high-level statements and which you cannot execute directly but only as part of a high-level statement. The ORDER BY and WHERE clauses are examples of this kind of clause. Finally, this section also includes some syntactically complex portions of statements called expressions, for example SelectExpression and TableSubquery . These clauses and expressions receive their own manual pages for ease of reference.
Unless it is explicitly stated otherwise, you can execute or prepare and then execute all the high-level statements, which are all marked with the word statement, via the interfaces provided by JDBC. This manual indicates whether an expression can be executed as a high-level statement.
The sections provide general information about statement use, and descriptions of the specific statements.
Interaction with the dependency system
Derby internally tracks the dependencies of prepared statements, which are SQL statements that are precompiled before being executed. Typically they are prepared (precompiled) once and executed multiple times.
Prepared statements depend on the dictionary objects and statements they reference. (Dictionary objects include tables, columns, constraints, indexes, views, and triggers.) Removing or modifying the dictionary objects or statements on which they depend invalidates them internally, which means that Derby will automatically try to recompile the statement when you execute it. If the statement fails to recompile, the execution request fails. However, if you take some action to restore the broken dependency (such as restoring the missing table), you can execute the same prepared statement, because Derby will recompile it automatically at the next execute request.
Statements depend on one another-an UPDATE WHERE CURRENT statement depends on the statement it references. Removing the statement on which it depends invalidates the UPDATE WHERE CURRENT statement.
In addition, prepared statements prevent execution of certain DDL statements if there are open results sets on them.
Manual pages for each statement detail what actions would invalidate that statement, if prepared.
Here is an example using the Derby tool ij:
ij> CREATE TABLE mytable (mycol INT); 0 rows inserted/updated/deleted ij> INSERT INTO mytable VALUES (1), (2), (3); 3 rows inserted/updated/deleted -- this example uses the ij command prepare, -- which prepares a statement ij> prepare p1 AS 'INSERT INTO MyTable VALUES (4)'; -- p1 depends on mytable; ij> execute p1; 1 row inserted/updated/deleted -- Derby executes it without recompiling ij> CREATE INDEX i1 ON mytable(mycol); 0 rows inserted/updated/deleted -- p1 is temporarily invalidated because of new index ij> execute p1; 1 row inserted/updated/deleted -- Derby automatically recompiles p1 and executes it ij> DROP TABLE mytable; 0 rows inserted/updated/deleted -- Derby permits you to drop table -- because result set of p1 is closed -- however, the statement p1 is temporarily invalidated ij> CREATE TABLE mytable (mycol INT); 0 rows inserted/updated/deleted ij> INSERT INTO mytable VALUES (1), (2), (3); 3 rows inserted/updated/deleted ij> execute p1; 1 row inserted/updated/deleted -- Because p1 is invalid, Derby tries to recompile it -- before executing. -- It is successful and executes. ij> DROP TABLE mytable; 0 rows inserted/updated/deleted -- statement p1 is now invalid, -- and this time the attempt to recompile it -- upon execution will fail ij> execute p1; ERROR 42X05: Table 'MYTABLE' does not exist.
ALTER TABLE statement
The ALTER TABLE statement allows you to:
 
add a column to a table
 
add a constraint to a table
 
drop an existing constraint from a table
 
increase the width of a VARCHAR, CHAR VARYING, and CHARACTER VARYING column
 
override row-level locking for the table (or drop the override)
Syntax
ALTER TABLE table-Name { ADD COLUMN column-definition | ADD CONSTRAINT clause | DROP { PRIMARY KEY | FOREIGN KEY constraint-name | UNIQUE constraint-name | CHECK constraint-name | CONSTRAINT constraint-name } ALTER column-alteration | LOCKSIZE { ROW | TABLE } }
column-definition
Simple-column-Name DataType [ Column-level-constraint ]* [ [ WITH ] DEFAULT {ConstantExpression | NULL } ]
column-alteration
column-Name SET DATA TYPE VARCHAR(integer) | column-name SET INCREMENT BY integer-constant
In the column-alteration, SET INCREMENT BY integer-constant, specifies the interval between consecutive values of the identity column. The next value to be generated for the identity column will be determined from the last assigned value with the increment applied. The column must already be defined with the IDENTITY attribute.
ALTER TABLE does not affect any view that references the table being altered. This includes views that have an "*" in their SELECT list. You must drop and re-create those views if you wish them to return the new columns.
Adding columns
The syntax for the column-definition for a new column is the same as for a column in a CREATE TABLE statement. This means that a column constraint can be placed on the new column within the ALTER TABLE ADD COLUMN statement. However, a column with a NOT NULL constraint can be added to an existing table if you give a default value; otherwise, an exception is thrown when the ALTER TABLE statement is executed.
Just as in CREATE TABLE, if the column definition includes a unique or primary key constraint, the column cannot contain null values, so the NOT NULL attribute must also be specified (SQLSTATE 42831).
Note: If a table has an UPDATE trigger without an explicit column list, adding a column to that table in effect adds that column to the implicit update column list upon which the trigger is defined, and all references to transition variables are invalidated so that they pick up the new column.
Adding constraints
ALTER TABLE ADD CONSTRAINT adds a table-level constraint to an existing table. Any supported table-level constraint type can be added via ALTER TABLE. The following limitations exist on adding a constraint to an existing table:
 
When adding a foreign key or check constraint to an existing table, Derby checks the table to make sure existing rows satisfy the constraint. If any row is invalid, Derby throws a statement exception and the constraint is not added.
 
All columns included in a primary key must contain non null data and be unique.
ALTER TABLE ADD UNIQUE or PRIMARY KEY provide a shorthand method of defining a primary key composed of a single column. If PRIMARY KEY is specified in the definition of column C, the effect is the same as if the PRIMARY KEY(C) clause were specified as a separate clause. The column cannot contain null values, so the NOT NULL attribute must also be specified.
For information on the syntax of constraints, see CONSTRAINT clause . Use the syntax for table-level constraint when adding a constraint with the ADD TABLE ADD CONSTRAINT syntax.
Dropping constraints
ALTER TABLE DROP CONSTRAINT drops a constraint on an existing table. To drop an unnamed constraint, you must specify the generated constraint name stored in SYS.SYSCONSTRAINTS as a delimited identifier.
Dropping a primary key, unique, or foreign key constraint drops the physical index that enforces the constraint (also known as a backing index).
Modifying columns
The column-alteration allows you to alter the named column in the following ways:
 
Increasing the length of an existing VARCHAR column. CHARACTER VARYING or CHAR VARYING can be used as synonyms for the VARCHAR keyword.
To increase the width of a column of these types, specify the data type and new size after the column name.
You are not allowed to decrease the width or to change the data type. You are not allowed to increase the width of a column that is part of a primary or unique key referenced by a foreign key constraint or that is part of a foreign key constraint.
 
Specifying the interval between consecutive values of the identity column.
To set an interval between consecutive values of the identity column, specify the integer-constant. You must previously define the column with the IDENTITY attribute (SQLSTATE 42837). If there are existing rows in the table, the values in the column for which the SET INCREMENT default was added do not change.
Setting defaults
You can specify a default value for a new column. A default value is the value that is inserted into a column if no other value is specified. If not explicitly specified, the default value of a column is NULL. If you add a default to a new column, existing rows in the table gain the default value in the new column.
For more information about defaults, see CREATE TABLE statement .
Changing the lock granularity for the table
The LOCKSIZE clause allows you to override row-level locking for the specific table, if your system uses the default setting of row-level locking. (If your system is set for table-level locking, you cannot change the locking granularity to row-level locking, although Derby allows you to use the LOCKSIZE clause in such a situation without throwing an exception.) To override row-level locking for the specific table, set locking for the table to TABLE. If you created the table with table-level locking granularity, you can change locking back to ROW with the LOCKSIZE clause in the ALTER TABLE STATEMENT. For information about why this is sometimes useful, see Tuning Derby .
Examples
-- Add a new column with a column-level constraint -- to an existing table -- An exception will be thrown if the table -- contains any rows -- since the newcol will be initialized to NULL -- in all existing rows in the table ALTER TABLE CITIES ADD COLUMN REGION VARCHAR(26) CONSTRAINT NEW_CONSTRAINT CHECK (REGION IS NOT NULL); -- Add a new unique constraint to an existing table -- An exception will be thrown if duplicate keys are found ALTER TABLE SAMP.DEPARTMENT ADD CONSTRAINT NEW_UNIQUE UNIQUE (DEPTNO); -- add a new foreign key constraint to the -- Cities table. Each row in Cities is checked -- to make sure it satisfied the constraints. -- if any rows don't satisfy the constraint, the -- constraint is not added ALTER TABLE CITIES ADD CONSTRAINT COUNTRY_FK Foreign Key (COUNTRY) REFERENCES COUNTRIES (COUNTRY); -- Add a primary key constraint to a table -- First, create a new table CREATE TABLE ACTIVITIES (CITY_ID INT NOT NULL, SEASON CHAR(2), ACTIVITY VARCHAR(32) NOT NULL); -- You will not be able to add this constraint if the -- columns you are including in the primary key have -- null data or duplicate values. ALTER TABLE Activities ADD PRIMARY KEY (city_id, activity); -- Drop a primary key constraint from the CITIES table ALTER TABLE Cities DROP CONSTRAINT Cities_PK; -- Drop a foreign key constraint from the CITIES table ALTER TABLE Cities DROP CONSTRAINT COUNTRIES_FK; -- add a DEPTNO column with a default value of 1 ALTER TABLE SAMP.EMP_ACT ADD COLUMN DEPTNO INT DEFAULT 1; -- increase the width of a VARCHAR column ALTER TABLE SAMP.EMP_PHOTO ALTER PHOTO_FORMAT SET DATA TYPE VARCHAR(30); -- change the lock granularity of a table ALTER TABLE SAMP.SALES LOCKSIZE TABLE;
Results
An ALTER TABLE statement causes all statements that are dependent on the table being altered to be recompiled before their next execution. ALTER TABLE is not allowed if there are any open cursors that reference the table being altered.
CREATE statements
Use the Create statements with functions, indexes, procedures, schemas, synonyms, tables, triggers, and views.
CREATE FUNCTION statement
The CREATE FUNCTION statement allows you to create Java functions, which you can then use in an expression.
Syntax
CREATE FUNCTION function-name ( [ FunctionParameter [, FunctionParameter ] ] * ) RETURNS DataType [ FunctionElement ] *
function-Name
If schema-Name is not provided, the current schema is the default schema. If a qualified procedure name is specified, the schema name cannot begin with SYS.
FunctionParameter
[ parameter-Name ] DataType
PararameterName must be unique within a function.
The syntax of DataType is described in Data types .
Note: Long data-types such as LONG VARCHAR, LONG VARCHAR FOR BIT DATA, CLOB, and BLOB are not allowed as parameters in a CREATE FUNCTION statement.
FunctionElement
{ | LANGUAGE { JAVA } | EXTERNAL NAME string | PARAMETER STYLE JAVA | { NO SQL | CONTAINS SQL | READS SQL DATA } | { RETURNS NULL ON NULL INPUT | CALLED ON NULL INPUT } }
LANGUAGE
JAVA- the database manager will call the function as a public static method in a Java class.
EXTERNAL NAME string
String describes the Java method to be called when the function is executed, and takes the following form:
class_name.method_name
The External Name cannot have any extraneous spaces.
PARAMETER STYLE
JAVA - The function will use a parameter-passing convention that conforms to the Java language and SQL Routines specification. INOUT and OUT parameters will be passed as single entry arrays to facilitate returning values. Result sets are returned through additional parameters to the Java method of type java.sql.ResultSet[] that are passed single entry arrays.
Derby does not support long column types (for example Long Varchar, BLOB, and so on). An error will occur if you try to use one of these long column types.
NO SQL, CONTAINS SQL, READS SQL DATA
Indicates whether the function issues any SQL statements and, if so, what type.
CONTAINS SQL
Indicates that SQL statements that neither read nor modify SQL data can be executed by the function. Statements that are not supported in any function return a different error.
NO SQL
Indicates that the function cannot execute any SQL statements
READS SQL DATA
Indicates that some SQL statements that do not modify SQL data can be included in the function. Statements that are not supported in any stored function return a different error. This is the default value.
RETURNS NULL ON NULL INPUT or CALLED ON NULL INPUT
Specifies whether the function is called if any of the input arguments is null. The result is the null value.
RETURNS NULL ON NULL INPUT
Specifies that the function is not invoked if any of the input arguments is null. The result is the null value.
CALLED ON NULL INPUT
Specifies that the function is invoked if any or all input arguments are null. This specification means that the function must be coded to test for null argument values. The function can return a null or non-null value. This is the default setting.
The function elements may appear in any order, but each type of element can only appear once. A function definition must contain these elements:
 
LANGUAGE
 
PARAMETER STYLE
 
EXTERNAL NAME
Example
CREATE FUNCTION TO_DEGREES(RADIANS DOUBLE) RETURNS DOUBLE PARAMETER STYLE JAVA NO SQL LANGUAGE JAVA EXTERNAL NAME 'java.lang.Math.toDegrees'
CREATE INDEX statement
A CREATE INDEX statement creates an index on a table. Indexes can be on one or more columns in the table.
Syntax
CREATE [UNIQUE] INDEX index-Name ON table-Name ( Simple-column-Name [ ASC | DESC ] [ , Simple-column-Name [ ASC | DESC ]] * )
The maximum number of columns for an index key in Derby is 16.
An index name cannot exceed 128 characters.
A column must not be named more than once in a single CREATE INDEX statement. Different indexes can name the same column, however.
Derby can use indexes to improve the performance of data manipulation statements (see Tuning Derby ). In addition, UNIQUE indexes provide a form of data integrity checking.
Index names are unique within a schema. (Some database systems allow different tables in a single schema to have indexes of the same name, but Derby does not.) Both index and table are assumed to be in the same schema if a schema name is specified for one of the names, but not the other. If schema names are specified for both index and table, an exception will be thrown if the schema names are not the same. If no schema name is specified for either table or index, the current schema is used.
By default, Derby uses the ascending order of each column to create the index. Specifying ASC after the column name does not alter the default behavior. The DESC keyword after the column name causes Derby to use descending order for the column to create the index. Using the descending order for a column can help improve the performance of queries that require the results in mixed sort order or descending order and for queries that select the minimum or maximum value of an indexed column.
If a qualified index name is specified, the schema name cannot begin with SYS.
Indexes and constraints
Unique, primary key, and foreign key constraints generate indexes that enforce or "back" the constraint (and are thus sometimes called backing indexes). If a column or set of columns has a UNIQUE or PRIMARY KEY constraint on it, you can not create an index on those columns. Derby has already created it for you with a system-generated name. System-generated names for indexes that back up constraints are easy to find by querying the system tables if you name your constraint. For example, to find out the name of the index that backs a constraint called FLIGHTS_PK:
SELECT CONGLOMERATENAME FROM SYS.SYSCONGLOMERATES, SYS.SYSCONSTRAINTS WHERE SYS.SYSCONGLOMERATES.TABLEID = SYSCONSTRAINTS.TABLEID AND CONSTRAINTNAME = 'FLIGHTS_PK'
CREATE INDEX OrigIndex ON Flights(orig_airport); -- money is usually ordered from greatest to least, -- so create the index using the descending order CREATE INDEX PAY_DESC ON SAMP.EMPLOYEE (SALARY); -- use a larger page size for the index call SYSCS_UTIL.SYSCS_SET_DATABASE_PROPERTY('derby.storage.pageSize','8192'); CREATE INDEX IXSALE ON SAMP.SALES (SALES); call SYSCS_UTIL.SYSCS_SET_DATABASE_PROPERTY('derby.storage.pageSize',NULL);
Page size and key size
Note: The size of the key columns in an index must be equal to or smaller than half the page size. If the length of the key columns in an existing row in a table is larger than half the page size of the index, creating an index on those key columns for the table will fail. This error only occurs when creating an index if an existing row in the table fails the criteria. After an index is created, inserts may fail if the size of their associated key exceeds the criteria.
Statement dependency system
Prepared statements that involve SELECT, INSERT, UPDATE, UPDATE WHERE CURRENT, DELETE, and DELETE WHERE CURRENT on the table referenced by the CREATE INDEX statement are invalidated when the index is created. Open cursors on the table are not affected.
CREATE PROCEDURE statement
The CREATE PROCEDURE statement allows you to create Java stored procedures, which you can then call using the CALL PROCEDURE statement.
Syntax
procedure-Name
If schema-Name is not provided, the current schema is the default schema. If a qualified procedure name is specified, the schema name cannot begin with SYS.
ProcedureParameter
[ { IN | OUT | INOUT } ] [ parameter-Name ] DataType
The default value for a parameter is IN. ParameterName must be unique within a procedure.
The syntax of DataType is described in Data types .
Note: Long data-types such as LONG VARCHAR, LONG VARCHAR FOR BIT DATA, CLOB, and BLOB are not allowed as parameters in a CREATE PROCEDURE statement.
ProcedureElement
{ | [ DYNAMIC ] RESULT SETS INTEGER | LANGUAGE { JAVA } | EXTERNAL NAME string | PARAMETER STYLE JAVA | { NO SQL | MODIFIES SQL DATA | CONTAINS SQL | READS SQL DATA } }
DYNAMIC RESULT SETS integer
Indicates the estimated upper bound of returned result sets for the procedure. Default is no (zero) dynamic result sets.
LANGUAGE
JAVA- the database manager will call the procedure as a public static method in a Java class.
EXTERNAL NAME string
String describes the Java method to be called when the procedure is executed, and takes the following form:
class_name.method_name
The External Name cannot have any extraneous spaces.
PARAMETER STYLE
JAVA - The procedure will use a parameter-passing convention that conforms to the Java language and SQL Routines specification. INOUT and OUT parameters will be passed as single entry arrays to facilitate returning values. Result sets are returned through additional parameters to the Java method of type java.sql.ResultSet [] that are passed single entry arrays.
Derby does not support long column types (for example Long Varchar, BLOB, and so on). An error will occur if you try to use one of these long column types.
NO SQL, CONTAINS SQL, READS SQL DATA, MODIFIES SQL DATA
Indicates whether the stored procedure issues any SQL statements and, if so, what type.
CONTAINS SQL
Indicates that SQL statements that neither read nor modify SQL data can be executed by the stored procedure. Statements that are not supported in any stored procedure return a different error. MODIFIES SQL DATA is the default value.
NO SQL
Indicates that the stored procedure cannot execute any SQL statements
READS SQL DATA
Indicates that some SQL statements that do not modify SQL data can be included in the stored procedure. Statements that are not supported in any stored procedure return a different error.
MODIFIES SQL DATA
Indicates that the stored procedure can execute any SQL statement except statements that are not supported in stored procedures.
The procedure elements may appear in any order, but each type of element can only appear once. A procedure definition must contain these elements:
 
LANGUAGE
 
PARAMETER STYLE
 
EXTERNAL NAME
Example
CREATE PROCEDURE SALES.TOTAL_REVENUE(IN S_MONTH INTEGER, IN S_YEAR INTEGER, OUT TOTAL DECIMAL(10,2)) PARAMETER STYLE JAVA READS SQL DATA LANGUAGE JAVA EXTERNAL NAME 'com.acme.sales.calculateRevenueByMonth'
CREATE SCHEMA statement
A schema is a way to logically group objects in a single collection and provide a unique namespace for objects.
Syntax
CREATE SCHEMA schemaName
The CREATE SCHEMA statement is used to create a schema. A schema name cannot exceed 128 characters. Schema names must be unique within the database.
-- Create a schema for employee-related tables CREATE SCHEMA EMP; -- Create a schema for airline-related tables CREATE SCHEMA Flights -- Create a table called "Availability" in each schema CREATE TABLE FLIGHTS.AVAILABILITY (FLIGHT_ID CHAR(6) NOT NULL, SEGMENT_NUMBER INT NOT NULL, FLIGHT_DATE DATE NOT NULL, ECONOMY_SEATS_TAKEN INT, BUSINESS_SEATS_TAKEN INT, FIRSTCLASS_SEATS_TAKEN INT, CONSTRAINT FLT_AVAIL_PK PRIMARY KEY (FLIGHT_ID, SEGMENT_NUMBER, FLIGHT_DATE)); CREATE TABLE EMP.AVAILABILITY (HOTEL_ID INT NOT NULL, BOOKING_DATE DATE NOT NULL, ROOMS_TAKEN INT, CONSTRAINT HOTELAVAIL_PK PRIMARY KEY (HOTEL_ID, BOOKING_DATE));
CREATE SYNONYM statement
Use the CREATE SYNONYM statement to provide an alternate name for a table or a view that is present in the same schema or another schema. You can also create synonyms for other synonyms, resulting in nested synonyms. A synonym can be used instead of the original qualified table or view name in SELECT, INSERT, UPDATE, DELETE or LOCK TABLE statements. You can create a synonym for a table or a view that doesn't exist, but the target table or view must be present before the synonym can be used.
Synonyms share the same namespace as tables or views. You cannot create a synonym with the same name as a table that already exists in the same schema. Similarly, you cannot create a table or view with a name that matches a synonym already present.
A synonym can be defined for a table/view that does not exist when you create the synonym. If the table or view doesn't exist, you will receive a warning message (SQLSTATE 01522). The referenced object must be present when you use a synonym in a DML statement.
You can create a nested synonym (a synonym for another synonym), but any attempt to create a synonym that results in a circular reference will return an error message (SQLSTATE 42916).
Synonyms cannot be defined in system schemas. All schemas starting with 'SYS' are considered system schemas and are reserved by Derby.
A synonym cannot be defined on a temporary table. Attempting to define a synonym on a temporary table will return an error message (SQLSTATE XCL51).
Syntax
CREATE SYNONYM synonym-Name FOR { view-Name | table-Name }
The synonym-Name in the statement represents the synonym name you are giving the target table or view, while the view-Name or table-Name represents the original name of the target table or view.
Example
CREATE SYNONYM SAMP.T1 FOR SAMP.TABLEWITHLONGNAME
CREATE TABLE statement
A CREATE TABLE statement creates a table. Tables contain columns and constraints, rules to which data must conform. Table-level constraints specify a column or columns. Columns have a data type and can specify column constraints (column-level constraints).
For information about constraints, see CONSTRAINT clause .
You can specify a default value for a column. A default value is the value to be inserted into a column if no other value is specified. If not explicitly specified, the default value of a column is NULL. See Column default .
You can specify storage properties such as page size for a table by calling the SYSCS_UTIL.SYSCS_SET_DATABASE_PROPERTY system procedure.
If a qualified table name is specified, the schema name cannot begin with SYS.
Example
CREATE TABLE HOTELAVAILABILITY (HOTEL_ID INT NOT NULL, BOOKING_DATE DATE NOT NULL, ROOMS_TAKEN INT DEFAULT 0, PRIMARY KEY (HOTEL_ID, BOOKING_DATE)); -- the table-level primary key definition allows you to -- include two columns in the primary key definition PRIMARY KEY (hotel_id, booking_date)) -- assign an identity column attribute to an INTEGER -- column, and also define a primary key constraint -- on the column CREATE TABLE PEOPLE (PERSON_ID INT NOT NULL GENERATED ALWAYS AS IDENTITY CONSTRAINT PEOPLE_PK PRIMARY KEY, PERSON VARCHAR(26)); -- assign an identity column attribute to a SMALLINT -- column with an initial value of 5 and an increment value -- of 5. CREATE TABLE GROUPS (GROUP_ID SMALLINT NOT NULL GENERATED ALWAYS AS IDENTITY (START WITH 5, INCREMENT BY 5), ADDRESS VARCHAR(100), PHONE VARCHAR(15));
Note: For more examples of CREATE TABLE statements using the various constraints, see CONSTRAINT clause .
column-definition:
Simple-column-Name DataType [ Column-level-constraint ]* [ [ WITH ] DEFAULT { ConstantExpression | NULL } | generated-column-spec ] [ Column-level-constraint ]*
The syntax of Data-Type is described in Data types .
Column default
For the definition of a default value, a ConstantExpression is an expression that does not refer to any table. It can include constants, date-time special registers, current schemas, users, and null.
generated-column-spec:
[ GENERATED { ALWAYS | BY DEFAULT } AS IDENTITY [ ( START WITH IntegerConstant [ ,INCREMENT BY IntegerConstant] ) ] ] ]
Identity column attributes
For SMALLINT, INT, and BIGINT columns with identity attributes, Derby automatically assigns increasing integer values to the column. Identity column attributes behave like other defaults in that when an insert statement does not specify a value for the column, Derby automatically provides the value. However, the value is not a constant; Derby automatically increments the default value at insertion time.
The IDENTITY keyword can only be specified if the data type associated with the column is one of the following exact integer types.
 
SMALLINT
 
INT
 
BIGINT
There are two kinds of identity columns in Derby: those which are GENERATED ALWAYS and those which are GENERATED BY DEFAULT.
GENERATED ALWAYS
An identity column that is GENERATED ALWAYS will increment the default value on every insertion and will store the incremented value into the column. Unlike other defaults, you cannot insert a value directly into or update an identity column that is GENERATED ALWAYS. Instead, either specify the DEFAULT keyword when inserting into the identity column, or leave the identity column out of the insertion column list altogether. For example:
create table greetings (i int generated always as identity, ch char(50)); insert into greetings values (DEFAULT, 'hello'); insert into greetings(ch) values ('bonjour');
Automatically generated values in a GENERATED ALWAYS identity column are unique. Creating an identity column does not create an index on the column.
GENERATED BY DEFAULT
An identity column that is GENERATED BY DEFAULT will only increment and use the default value on insertions when no explicit value is given. Unlike GENERATED ALWAYS columns, you can specify a particular value in an insertion statement to be used instead of the generated default value.
To use the generated default, either specify the DEFAULT keyword when inserting into the identity column, or just leave the identity column out of the insertion column list. To specify a value, included it in the insertion statement. For example:
create table greetings (i int generated by default as identity, ch char(50)); -- specify value "1": insert into greetings values (1, 'hi'); -- use generated default insert into greetings values (DEFAULT, 'salut'); -- use generated default insert into greetings(ch) values ('bonjour');
Note that unlike a GENERATED ALWAYS column, a GENERATED BY DEFAULT column does not guarantee uniqueness. Thus, in the above example, the hi and salut rows will both have an identity value of "1", because the generated column starts at "1" and the user-specified value was also "1". To prevent duplication, especially when loading or importing data, create the table using the START WITH value which corresponds to the first identity value that the system should assign. To check for this condition and disallow it, you can use a primary key or unique constraint on the GENERATED BY DEFAULT identity column.
By default, the initial value of an identity column is 1, and the amount of the increment is 1. You can specify non-default values for both the initial value and the interval amount when you define the column with the key words STARTS WITH and INCREMENT BY. And if you specify a negative number for the increment value, Derby decrements the value with each insert. If this value is 0, or positive, Derby increments the value with each insert.
The maximum and minimum values allowed in identity columns are determined by the data type of the column. Attempting to insert a value outside the range of values supported by the data type raises an exception.
Table1. Maximum and Minimum Values for Columns with Generated Column Specs
Data type
Maximum Value
Minimum Value
SMALLINT
32767 (java.lang.Short.MAX_VALUE)
-32768 (java.lang.Short.MIN_VALUE)
INT
2147483647 (java.lang.Integer.MAX_VALUE)
-2147483648 (java.lang.Integer.MIN_VALUE)
BIGINT
9223372036854775807 (java.lang.Long.MAX_VALUE)
-9223372036854775808 (java.lang.Long.MIN_VALUE)
Automatically generated values in an identity column are unique. Use a primary key or unique constraint on a column to guarantee uniqueness. Creating an identity column does not create an index on the column.
The IDENTITY_VAL_LOCAL function is a non-deterministic function that returns the most recently assigned value for an identity column. See IDENTITY_VAL_LOCAL for more information.
Note: Specify the schema, table, and column name using the same case as those names are stored in the system tables--that is, all upper case unless you used delimited identifiers when creating those database objects.
Derby keeps track of the last increment value for a column in a cache. It also stores the value of what the next increment value will be for the column on disk in the AUTOINCREMENTVALUE column of the SYS.SYSCOLUMNS system table. Rolling back a transaction does not undo this value, and thus rolled-back transactions can leave "gaps" in the values automatically inserted into an identity column. Derby behaves this way to avoid locking a row in SYS.SYSCOLUMNS for the duration of a transaction and keeping concurrency high.
When an insert happens within a triggered-SQL-statement, the value inserted by the triggered-SQL-statement into the identity column is available from ConnectionInfo only within the trigger code. The trigger code is also able to see the value inserted by the statement that caused the trigger to fire. However, the statement that caused the trigger to fire is not able to see the value inserted by the triggered-SQL-statement into the identity column. Likewise, triggers can be nested (or recursive). An SQL statement can cause trigger T1 to fire. T1 in turn executes an SQL statement that causes trigger T2 to fire. If both T1 and T2 insert rows into a table that cause Derby to insert into an identity column, trigger T1 cannot see the value caused by T2's insert, but T2 can see the value caused by T1's insert. Each nesting level can see increment values generated by itself and previous nesting levels, all the way to the top-level SQL statement that initiated the recursive triggers. You can only have 16 levels of trigger recursion.
Example
create table greetings (i int generated by default as identity (START WITH 2, INCREMENT BY 1), ch char(50)); -- specify value "1": insert into greetings values (1, 'hi'); -- use generated default insert into greetings values (DEFAULT, 'salut'); -- use generated default insert into greetings(ch) values ('bonjour');
CREATE TRIGGER statement
A trigger defines a set of actions that are executed when a database event occurs on a specified table. A database event is a delete, insert, or update operation. For example, if you define a trigger for a delete on a particular table, the trigger's action occurs whenever someone deletes a row or rows from the table.
Along with constraints, triggers can help enforce data integrity rules with actions such as cascading deletes or updates. Triggers can also perform a variety of functions such as issuing alerts, updating other tables, sending e-mail, and other useful actions.
You can define any number of triggers for a single table, including multiple triggers on the same table for the same event.
You can create a trigger in any schema except one that starts with SYS. The trigger need not reside in the same schema as the table on which it is defined.
If a qualified trigger name is specified, the schema name cannot begin with SYS.
Syntax
CREATE TRIGGER TriggerName { AFTER | NO CASCADE BEFORE } { INSERT | DELETE | UPDATE } [ OF column-Name [, column-Name ]* ] ON table-Name [ ReferencingClause ] FOR EACH { ROW | STATEMENT } MODE DB2SQL Triggered-SQL-statement
Before or after: when triggers fire
Triggers are defined as either Before or After triggers.
 
Before triggers fire before the statement's changes are applied and before any constraints have been applied. Before triggers can be either row or statement triggers (see Statement versus row triggers ).
 
After triggers fire after all constraints have been satisfied and after the changes have been applied to the target table. After triggers can be either row or statement triggers (see Statement versus row triggers ).
Insert, delete, or update: what causes the trigger to fire
A trigger is fired by one of the following database events, depending on how you define it (see Syntax above):
 
INSERT
 
UPDATE
 
DELETE
You can define any number of triggers for a given event on a given table. For update, you can specify columns.
Referencing old and new values: the referencing clause
Many triggered-SQL-statements need to refer to data that is currently being changed by the database event that caused them to fire. The triggered-SQL-statement might need to refer to the new (post-change or "after") values.
Derby provides you with a number of ways to refer to data that is currently being changed by the database event that caused the trigger to fire. Changed data can be referred to in the triggered-SQL-statement using transition variables or transition tables. The referencing clause allows you to provide a correlation name or alias for these transition variables by specifying OLD/NEW AS correlation-Name .
For example, if you add the following clause to the trigger definition:
REFERENCING OLD AS DELETEDROW
you can then refer to this correlation name in the triggered-SQL-statement:
DELETE FROM HotelAvailability WHERE hotel_id = DELETEDROW.hotel_id
The OLD and NEW transition variables map to a java.sql.ResultSet with a single row.
Note: Only row triggers (see Statement versus row triggers ) can use the transition variables. INSERT row triggers cannot reference an OLD row. DELETE row triggers cannot reference a NEW row.
For statement triggers, transition tables serve as a table identifier for the triggered-SQL-statement or the trigger qualification. The referencing clause allows you to provide a correlation name or alias for these transition tables by specifying OLD_TABLE/NEW_TABLE AS correlation-Name
For example:
REFERENCING OLD_TABLE AS DeletedHotels
allows you to use that new identifier (DeletedHotels) in the triggered-SQL-statement:
DELETE FROM HotelAvailability WHERE hotel_id IN (SELECT hotel_id FROM DeletedHotels)
The old and new transition tables map to a java.sql.ResultSet with cardinality equivalent to the number of rows affected by the triggering event.
Note: Only statement triggers (see Statement versus row triggers ) can use the transition tables. INSERT statement triggers cannot reference an OLD table. DELETE statement triggers cannot reference a NEW table.
The referencing clause can designate only one new correlation or identifier and only one old correlation or identifier. Row triggers cannot designate an identifier for a transition table and statement triggers cannot designate a correlation for transition variables.
Statement versus row triggers
You must specify whether a trigger is a statement trigger or a row trigger:
 
statement triggers
A statement trigger fires once per triggering event and regardless of whether any rows are modified by the insert, update, or delete event.
 
row triggers
A row trigger fires once for each row affected by the triggering event. If no rows are affected, the trigger does not fire.
Note: An update that sets a column value to the value that it originally contained (for example, UPDATE T SET C = C) causes a row trigger to fire, even though the value of the column is the same as it was prior to the triggering event.
Triggered-SQL-statement
The action defined by the trigger is called the triggered-SQL-statement (in Syntax above, see the last line). It has the following limitations:
 
It must not contain any dynamic parameters (?).
 
It must not create, alter, or drop the table upon which the trigger is defined.
 
It must not add an index to or remove an index from the table on which the trigger is defined.
 
It must not add a trigger to or drop a trigger from the table upon which the trigger is defined.
 
It must not commit or roll back the current transaction or change the isolation level.
 
It must not execute a CALL statement.
 
Before triggers cannot have INSERT, UPDATE or DELETE statements as their action.
The triggered-SQL-statement can reference database objects other than the table upon which the trigger is declared. If any of these database objects is dropped, the trigger is invalidated. If the trigger cannot be successfully recompiled upon the next execution, the invocation throws an exception and the statement that caused it to fire will be rolled back.
For more information on triggered-SQL-statements, see the Derby Developer's Guide .
Order of execution
When a database event occurs that fires a trigger, Derby performs actions in this order:
 
It fires No Cascade Before triggers.
 
It performs constraint checking (primary key, unique key, foreign key, check).
 
It performs the insert, update, or delete.
 
It fires After triggers.
When multiple triggers are defined for the same database event for the same table for the same trigger time (before or after), triggers are fired in the order in which they were created.
-- Statements and triggers: CREATE TRIGGER t1 NO CASCADE BEFORE UPDATE ON x FOR EACH ROW MODE DB2SQL values app.notifyEmail('Jerry', 'Table x is about to be updated'); CREATE TRIGGER FLIGHTSDELETE AFTER DELETE ON FLIGHTS REFERENCING OLD_TABLE AS DELETEDFLIGHTS FOR EACH STATEMENT MODE DB2SQL DELETE FROM FLIGHTAVAILABILITY WHERE FLIGHT_ID IN (SELECT FLIGHT_ID FROM DELETEDFLIGHTS); CREATE TRIGGER FLIGHTSDELETE3 AFTER DELETE ON FLIGHTS REFERENCING OLD AS OLD FOR EACH ROW MODE DB2SQL DELETE FROM FLIGHTAVAILABILITY WHERE FLIGHT_ID = OLD.FLIGHT_ID;
Note: You can find more examples in the Derby Developer's Guide .
Trigger recursion
The maximum trigger recursion depth is 16.
Related information
Special system functions that return information about the current time or current user are evaluated when the trigger fires, not when it is created. Such functions include:
 
 
 
 
 
 
ReferencingClause:
REFERENCING { { OLD | NEW } [ AS ] correlation-Name [ { OLD | NEW } [ AS ] correlation-Name ] | { OLD_TABLE | NEW_TABLE } [ AS ] Identifier [ { OLD_TABLE | NEW_TABLE } [AS] Identifier ] }
CREATE VIEW statement
Views are virtual tables formed by a query. A view is a dictionary object that you can use until you drop it.
Views are not updatable.
If a qualified view name is specified, the schema name cannot begin with SYS.
Syntax
CREATE VIEW view-Name [ ( Simple-column-Name [, Simple-column-Name ] * ) ] AS Query
A view definition can contain an optional view column list to explicitly name the columns in the view. If there is no column list, the view inherits the column names from the underlying query. All columns in a view must be uniquely named.
CREATE VIEW SAMP.V1 (COL_SUM, COL_DIFF) AS SELECT COMM + BONUS, COMM - BONUS FROM SAMP.EMPLOYEE; CREATE VIEW SAMP.VEMP_RES (RESUME) AS VALUES 'Delores M. Quintana', 'Heather A. Nicholls', 'Bruce Adamson'; CREATE VIEW SAMP.PROJ_COMBO (PROJNO, PRENDATE, PRSTAFF, MAJPROJ) AS SELECT PROJNO, PRENDATE, PRSTAFF, MAJPROJ FROM SAMP.PROJECT UNION ALL SELECT PROJNO, EMSTDATE, EMPTIME, EMPNO FROM SAMP.EMP_ACT WHERE EMPNO IS NOT NULL;
Statement dependency system
View definitions are dependent on the tables and views referenced within the view definition. DML (data manipulation language) statements that contain view references depend on those views, as well as the objects in the view definitions that the views are dependent on. Statements that reference the view depend on indexes the view uses; which index a view uses can change from statement to statement based on how the query is optimized. For example, given:
CREATE TABLE T1 (C1 DOUBLE PRECISION); CREATE FUNCTION SIN (DATA DOUBLE) RETURNS DOUBLE EXTERNAL NAME 'java.lang.Math.sin' LANGUAGE JAVA PARAMETER STYLE JAVA; CREATE VIEW V1 (C1) AS SELECT SIN(C1) FROM T1;
the following SELECT:
SELECT * FROM V1
is dependent on view V1, table T1, and external scalar function SIN.
DROP Statements
Use Drop statements with functions, indexes, procedures, schemas, synonyms, tables, triggers, and views.
DROP FUNCTION statement
Syntax
DROP FUNCTION function-name
Identifies the particular function to be dropped, and is valid only if there is exactly one function instance with the function-name in the schema. The identified function can have any number of parameters defined for it. If no function with the indicated name in the named or implied schema, an error (SQLSTATE 42704) will occur. An error will also occur if there is more than one specific instance of the function in the named or implied schema.
DROP INDEX statement
DROP INDEX removes the specified index.
Syntax
DROP INDEX index-Name
DROP INDEX OrigIndex DROP INDEX DestIndex
Statement dependency system
If there is an open cursor on the table from which the index is dropped, the DROP INDEX statement generates an error and does not drop the index. Otherwise, statements that depend on the index's table are invalidated.
DROP PROCEDURE statement
Syntax
DROP PROCEDURE procedure-Name
Identifies the particular procedure to be dropped, and is valid only if there is exactly one procedure instance with the procedure-name in the schema. The identified procedure can have any number of parameters defined for it. If no procedure with the indicated name in the named or implied schema, an error (SQLSTATE 42704) will occur. An error will also occur if there is more than one specific instance of the procedure in the named or implied schema.
DROP SCHEMA statement
The DROP SCHEMA statement drops a schema. The target schema must be empty for the drop to succeed.
Neither the APP schema (the default user schema) nor the SYS schema can be dropped.
Syntax
DROP SCHEMA schemaName RESTRICT
The RESTRICT keyword enforces the rule that no objects can be defined in the specified schema for the schema to be deleted from the database. The RESTRICT keyword is required
-- Drop the SAMP schema -- The SAMP schema may only be deleted from the database -- if no objects are defined in the SAMP schema. DROP SCHEMA SAMP RESTRICT
DROP SYNONYM statement
Drops the specified synonym from a table or view.
Syntax
DROP SYNONYM synonym-Name
DROP TABLE statement
DROP TABLE removes the specified table.
Syntax
DROP TABLE table-Name
Statement dependency system
Triggers, constraints (primary, unique, check and references from the table being dropped) and indexes on the table are silently dropped. The existence of an open cursor that references table being dropped cause the DROP TABLE statement to generate an error, and the table is not dropped.
Dropping a table invalidates statements that depend on the table. (Invalidating a statement causes it to be recompiled upon the next execution. See Interaction with the dependency system .)
DROP TRIGGER statement
DROP TRIGGER removes the specified trigger.
Syntax
DROP TRIGGER TriggerName
DROP TRIGGER TRIG1
Statement dependency system
When a table is dropped, all triggers on that table are automatically dropped. (You don't have to drop a table's triggers before dropping the table.)
DROP VIEW statement
Drops the specified view.
Syntax
DROP VIEW view-Name
DROP VIEW AnIdentifier
Statement dependency system
Any statements referencing the view are invalidated on a DROP VIEW statement. DROP VIEW is disallowed if there are any views or open cursors dependent on the view. The view must be dropped before any objects that it is dependent on can be dropped.
RENAME statements
Use the Rename statements with indexes and tables.
RENAME INDEX statement
This statement allows you to rename an index in the current schema. Users cannot rename indexes in the SYS schema.
Syntax
RENAME INDEX index-Name TO new-index-Name
RENAME INDEX DESTINDEX TO ARRIVALINDEX
Statement dependency system
RENAME INDEX is not allowed if there are any open cursors that reference the index being renamed.
RENAME TABLE statement
RENAME TABLE allows you to rename an existing table in any schema (except the schema SYS).
Syntax
RENAME TABLE table-Name TO new-Table-Name
If there is a view or foreign key that references the table, attempts to rename it will generate an error. In addition, if there are any check constraints or triggers on the table, attempts to rename it will also generate an error.
RENAME TABLE SAMP.EMP_ACT TO EMPLOYEE_ACT
Also see ALTER TABLE statement for more information.
Statement dependency system
If there is an index defined on the table, the table can be renamed.
RENAME TABLE is not allowed if there are any open cursors that reference the table being altered.
SET statements
Use the Set statements with schemas and to set the current isolation level.
SET SCHEMA statement
The SET SCHEMA statement sets the default schema for a connection's session to the designated schema. The default schema is used as the target schema for all statements issued from the connection that do not explicitly specify a schema name.
The target schema must exist for the SET SCHEMA statement to succeed. If the schema doesn't exist an error is returned. See CREATE SCHEMA statement .
The SET SCHEMA statement is not transactional: If the SET SCHEMA statement is part of a transaction that is rolled back, the schema change remains in effect.
Syntax
SET [CURRENT] SCHEMA [=] { schemaName | USER | ? | '<string-constant>' } | SET CURRENT SQLID [=] { schemaName | USER | ? | '<string-constant>' }
schemaName is an identifier with a maximum length of 128. It is case insensitive unless enclosed in double quotes. (For example, SYS is equivalent to sYs, SYs, sys, etcetera.)
USER is the current user. If no current user is defined, the current schema defaults the APP schema. (If a user name was specified upon connection, the user's name is the default schema for the connection, if a schema with that name exists.)
? is a dynamic parameter specification that can be used in prepared statements. The SET SCHEMA statement can be prepared once and then executed with different schema values. The schema values are treated as string constants so they are case sensitive. For example, to designate the APP schema, use the string "APP" rather than "app".
-- the following are all equivalent and will work -- assuming a schema called HOTEL SET SCHEMA HOTEL SET SCHEMA hotel SET CURRENT SCHEMA hotel SET CURRENT SQLID hotel SET SCHEMA = hotel SET CURRENT SCHEMA = hotel SET CURRENT SQLID = hotel SET SCHEMA "HOTEL" -- quoted identifier SET SCHEMA 'HOTEL' -- quoted string --This example produces an error because --lower case hotel won't be found SET SCHEMA = 'hotel' --This example produces an error because SQLID is not --allowed without CURRENT SET SQLID hotel -- This sets the schema to the current user id SET CURRENT SCHEMA USER // Here's an example of using set schema in an Java program PreparedStatement ps = conn.PrepareStatement("set schema ?"); ps.setString(1,"HOTEL"); ps.executeUpdate(); ... do some work ps.setString(1,"APP"); ps.executeUpdate(); ps.setString(1,"app"); //error - string is case sensitive // no app will be found ps.setNull(1, Types.VARCHAR); //error - null is not allowed
SET CURRENT ISOLATION statement
The SET CURRENT ISOLATION LEVEL statement allows a user to change the isolation level for the user's connection. Valid levels are SERIALIZABLE, REPEATABLE READ, READ COMMITTED, and READ UNCOMMITTED.
Issuing this command commits the current transaction, which is consistent with the java.sql.Connection.setTransactionLevel method.
For information about isolation levels, see "Locking, Concurrency, and Isolation" in the Derby Developer's Guide .
Syntax
SET [ CURRENT ] ISOLATION [ = ] { UR | DIRTY READ | READ UNCOMMITTED CS | READ COMMITTED | CURSOR STABILITY RS | RR | REPEATABLE READ | SERIALIZABLE RESET }
set isolation serializable;
CALL (PROCEDURE)
The CALL (PROCEDURE) statement is used to call procedures. A call to a procedure does not return any value.
Syntax
CALL procedure-Name ( [ ? [, ?]* ] )
Example
CREATE PROCEDURE SALES.TOTAL_REVENUE(IN S_MONTH INTEGER, IN S_YEAR INTEGER, OUT TOTAL DECIMAL(10,2)) PARAMETER STYLE JAVA READS SQL DATA LANGUAGE JAVA EXTERNAL NAME 'com.acme.sales.calculateRevenueByMonth'; CALL SALES.TOTAL_REVENUE(?,?,?);
CONSTRAINT clause
A CONSTRAINT clause is an optional part of a CREATE TABLE statement or ALTER TABLE statement . A constraint is a rule to which data must conform. Constraint names are optional.
A CONSTRAINT can be one of the following:
 
a column-level constraint
Column-level constraints refer to a single column in the table and do not specify a column name (except check constraints). They refer to the column that they follow.
 
a table-level constraint
Table-level constraints refer to one or more columns in the table. Table-level constraints specify the names of the columns to which they apply. Table-level CHECK constraints can refer to 0 or more columns in the table.
Column constraints include:
 
NOT NULL
Specifies that this column cannot hold NULL values (constraints of this type are not nameable).
 
PRIMARY KEY
Specifies the column that uniquely identifies a row in the table. The identified columns must be defined as NOT NULL.
Note: If you attempt to add a primary key using ALTER TABLE and any of the columns included in the primary key contain null values, an error will be generated and the primary key will not be added. See ALTER TABLE statement for more information.
 
UNIQUE
Specifies that values in the column must be unique. NULL values are not allowed.
 
FOREIGN KEY
Specifies that the values in the column must correspond to values in a referenced primary key or unique key column or that they are NULL.
 
CHECK
Specifies rules for values in the column.
Table constraints include:
 
PRIMARY KEY
Specifies the column or columns that uniquely identify a row in the table. NULL values are not allowed.
 
UNIQUE
Specifies that values in the columns must be unique. The identified columns must be defined as NOT NULL.
 
FOREIGN KEY
Specifies that the values in the columns must correspond to values in referenced primary key or unique columns or that they are NULL.
Note: If the foreign key consists of multiple columns, and any column is NULL, the whole key is considered NULL. The insert is permitted no matter what is on the non-null columns.
 
CHECK
Specifies a wide range of rules for values in the table.
Column constraints and table constraints have the same function; the difference is in where you specify them. Table constraints allow you to specify more than one column in a PRIMARY KEY, UNIQUE, CHECK, or FOREIGN KEY constraint definition. Column-level constraints (except for check constraints) refer to only one column.
Syntax
Primary key and unique constraints
A primary key defines the set of columns that uniquely identifies rows in a table.
When you create a primary key constraint, none of the columns included in the primary key can have NULL constraints; that is, they must not permit NULL values.
ALTER TABLE ADD PRIMARY KEY allows you to include existing columns in a primary key if they were first defined as NOT NULL. NULL values are not allowed. If the column(s) contain NULL values, the system will not add the primary key constraint. See ALTER TABLE statement for more information.
A table can have at most one PRIMARY KEY constraint, but can have multiple UNIQUE constraints.
Foreign key constraints
Foreign keys provide a way to enforce the referential integrity of a database. A foreign key is a column or group of columns within a table that references a key in some other table (or sometimes, though rarely, the same table). The foreign key must always include the columns of which the types exactly match those in the referenced primary key or unique constraint.
For a table-level foreign key constraint in which you specify the columns in the table that make up the constraint, you cannot use the same column more than once.
If there is a column list in the ReferencesSpecification (a list of columns in the referenced table), it must correspond either to a unique constraint or to a primary key constraint in the referenced table. The ReferencesSpecification can omit the column list for the referenced table if that table has a declared primary key.
If there is no column list in the ReferencesSpecification and the referenced table has no primary key, a statement exception is thrown. (This means that if the referenced table has only unique keys, you must include a column list in the ReferencesSpecification.)
A foreign key constraint is satisfied if there is a matching value in the referenced unique or primary key column. If the foreign key consists of multiple columns, the foreign key value is considered NULL if any of its columns contains a NULL.
Note: It is possible for a foreign key consisting of multiple columns to allow one of the columns to contain a value for which there is no matching value in the referenced columns, per the SQL-92 standard. To avoid this situation, create NOT NULL constraints on all of the foreign key's columns.
Foreign key constraints and DML
When you insert into or update a table with an enabled foreign key constraint, Derby checks that the row does not violate the foreign key constraint by looking up the corresponding referenced key in the referenced table. If the constraint is not satisfied, Derby rejects the insert or update with a statement exception.
When you update or delete a row in a table with a referenced key (a primary or unique constraint referenced by a foreign key), Derby checks every foreign key constraint that references the key to make sure that the removal or modification of the row does not cause a constraint violation. If removal or modification of the row would cause a constraint violation, the update or delete is not permitted and Derby throws a statement exception.
Derby performs constraint checks at the time the statement is executed, not when the transaction commits.
Backing indexes
UNIQUE, PRIMARY KEY, and FOREIGN KEY constraints generate indexes that enforce or "back" the constraint (and are sometimes called backing indexes). UNIQUE and PRIMARY KEY constraints generate unique indexes. FOREIGN KEY constraints generate non-unique indexes. Therefore, if a column or set of columns has a UNIQUE, PRIMARY KEY, or FOREIGN KEY constraint on it, you do not need to create an index on those columns for performance. Derby has already created it for you. See Indexes and constraints .
These indexes are available to the optimizer for query optimization (see CREATE INDEX statement ) and have system-generated names.
You cannot drop backing indexes with a DROP INDEX statement; you must drop the constraint or the table.
Check constraints
A check constraint can be used to specify a wide range of rules for the contents of a table. A search condition (which is a boolean expression) is specified for a check constraint. This search condition must be satisfied for all rows in the table. The search condition is applied to each row that is modified on an INSERT or UPDATE at the time of the row modification. The entire statement is aborted if any check constraint is violated.
Requirements for search condition
If a check constraint is specified as part of a column-definition, a column reference can only be made to the same column. Check constraints specified as part of a table definition can have column references identifying columns previously defined in the CREATE TABLE statement.
The search condition must always return the same value if applied to the same values. Thus, it cannot contain any of the following:
 
Dynamic parameters (?)
 
Date/Time Functions (CURRENT_DATE, CURRENT_TIME, CURRENT_TIMESTAMP)
 
Subqueries
 
User Functions (such as USER, SESSION_USER, CURRENT_USER)
Referential actions
You can specify an ON DELETE clause and/or an ON UPDATE clause, followed by the appropriate action (CASCADE, RESTRICT, SET NULL, or NO ACTION) when defining foreign keys. These clauses specify whether Derby should modify corresponding foreign key values or disallow the operation, to keep foreign key relationships intact when a primary key value is updated or deleted from a table.
You specify the update and delete rule of a referential constraint when you define the referential constraint.
The update rule applies when a row of either the parent or dependent table is updated. The choices are NO ACTION and RESTRICT.
When a value in a column of the parent table's primary key is updated and the update rule has been specified as RESTRICT, Derby checks dependent tables for foreign key constraints. If any row in a dependent table violates a foreign key constraint, the transaction is rolled back.
If the update rule is NO ACTION, Derby checks the dependent tables for foreign key constraints after all deletes have been executed but before triggers have been executed. If any row in a dependent table violates a foreign key constraint, the statement is rejected.
When a value in a column of the dependent table is updated, and that value is part of a foreign key, NO ACTION is the implicit update rule. NO ACTION means that if a foreign key is updated with a non-null value, the update value must match a value in the parent table's primary key when the update statement is completed. If the update does not match a value in the parent table's primary key, the statement is rejected.
The delete rule applies when a row of the parent table is deleted and that row has dependents in the dependent table of the referential constraint. If rows of the dependent table are deleted, the delete operation on the parent table is said to be propagated to the dependent table. If the dependent table is also a parent table, the action specified applies, in turn, to its dependents.
The choices are NO ACTION, RESTRICT, CASCADE, or SET NULL. SET NULL can be specified only if some column of the foreign key allows null values.
If the delete rule is:
NO ACTION, Derby checks the dependent tables for foreign key constraints after all deletes have been executed but before triggers have been executed. If any row in a dependent table violates a foreign key constraint, the statement is rejected.
RESTRICT, Derby checks dependent tables for foreign key constraints. If any row in a dependent table violates a foreign key constraint, the transaction is rolled back.
CASCADE, the delete operation is propagated to the dependent table (and that table's dependents, if applicable).
SET NULL, each nullable column of the dependent table's foreign key is set to null. (Again, if the dependent table also has dependent tables, nullable columns in those tables' foreign keys are also set to null.)
Each referential constraint in which a table is a parent has its own delete rule; all applicable delete rules are used to determine the result of a delete operation. Thus, a row cannot be deleted if it has dependents in a referential constraint with a delete rule of RESTRICT or NO ACTION. Similarly, a row cannot be deleted if the deletion cascades to any of its descendants that are dependents in a referential constraint with the delete rule of RESTRICT or NO ACTION.
Deleting a row from the parent table involves other tables. Any table involved in a delete operation on the parent table is said to be delete-connected to the parent table. The delete can affect rows of these tables in the following ways:
 
If the delete rule is RESTRICT or NO ACTION, a dependent table is involved in the operation but is not affected by the operation. (That is, Derby checks the values within the table, but does not delete any values.)
 
If the delete rule is SET NULL, a dependent table's rows can be updated when a row of the parent table is the object of a delete or propagated delete operation.
 
If the delete rule is CASCADE, a dependent table's rows can be deleted when a parent table is the object of a delete.
 
If the dependent table is also a parent table, the actions described in this list apply, in turn, to its dependents.
Examples
-- column-level primary key constraint named OUT_TRAY_PK: CREATE TABLE SAMP.OUT_TRAY ( SENT TIMESTAMP, DESTINATION CHAR(8), SUBJECT CHAR(64) NOT NULL CONSTRAINT OUT_TRAY_PK PRIMARY KEY, NOTE_TEXT VARCHAR(3000) ); -- the table-level primary key definition allows you to -- include two columns in the primary key definition: CREATE TABLE SAMP.SCHED ( CLASS_CODE CHAR(7) NOT NULL, DAY SMALLINT NOT NULL, STARTING TIME, ENDING TIME, PRIMARY KEY (CLASS_CODE, DAY) ); -- Use a column-level constraint for an arithmetic check -- Use a table-level constraint -- to make sure that a employee's taxes does not -- exceed the bonus CREATE TABLE SAMP.EMP ( EMPNO CHAR(6) NOT NULL CONSTRAINT EMP_PK PRIMARY KEY, FIRSTNME CHAR(12) NOT NULL, MIDINIT vARCHAR(12) NOT NULL, LASTNAME VARCHAR(15) NOT NULL, SALARY DECIMAL(9,2) CONSTRAINT SAL_CK CHECK (SALARY >= 10000), BONUS DECIMAL(9,2), TAX DECIMAL(9,2), CONSTRAINT BONUS_CK CHECK (BONUS > TAX) ); -- use a check constraint to allow only appropriate -- abbreviations for the meals CREATE TABLE FLIGHTS ( FLIGHT_ID CHAR(6) NOT NULL , SEGMENT_NUMBER INTEGER NOT NULL , ORIG_AIRPORT CHAR(3), DEPART_TIME TIME, DEST_AIRPORT CHAR(3), ARRIVE_TIME TIME, MEAL CHAR(1) CONSTRAINT MEAL_CONSTRAINT CHECK (MEAL IN ('B', 'L', 'D', 'S')), PRIMARY KEY (FLIGHT_ID, SEGMENT_NUMBER) ); CREATE TABLE METROPOLITAN ( HOTEL_ID INT NOT NULL CONSTRAINT HOTELS_PK PRIMARY KEY, HOTEL_NAME VARCHAR(40) NOT NULL, CITY_ID INT CONSTRAINT METRO_FK REFERENCES CITIES ); -- create a table with a table-level primary key constraint -- and a table-level foreign key constraint CREATE TABLE FLTAVAIL ( FLIGHT_ID CHAR(6) NOT NULL, SEGMENT_NUMBER INT NOT NULL, FLIGHT_DATE DATE NOT NULL, ECONOMY_SEATS_TAKEN INT, BUSINESS_SEATS_TAKEN INT, FIRSTCLASS_SEATS_TAKEN INT, CONSTRAINT FLTAVAIL_PK PRIMARY KEY (FLIGHT_ID, SEGMENT_NUMBER), CONSTRAINT FLTS_FK FOREIGN KEY (FLIGHT_ID, SEGMENT_NUMBER) REFERENCES Flights (FLIGHT_ID, SEGMENT_NUMBER) ); -- add a unique constraint to a column ALTER TABLE SAMP.PROJECT ADD CONSTRAINT P_UC UNIQUE (PROJNAME); -- create a table whose city_id column references the -- primary key in the Cities table -- using a column-level foreign key constraint CREATE TABLE CONDOS ( CONDO_ID INT NOT NULL CONSTRAINT hotels_PK PRIMARY KEY, CONDO_NAME VARCHAR(40) NOT NULL, CITY_ID INT CONSTRAINT city_foreign_key REFERENCES Cities ON DELETE CASCADE ON UPDATE RESTRICT );
Statement dependency system
INSERT and UPDATE statements depend on all constraints on the target table. DELETEs depend on unique, primary key, and foreign key constraints. These statements are invalidated if a constraint is added to or dropped from the target table.
Column-level-constraint
{ NOT NULL | [ [CONSTRAINT constraint-Name ] { CHECK ( searchCondition ) | { PRIMARY KEY | UNIQUE | REFERENCES clause } } }
Table-level constraint
[CONSTRAINT constraint-Name ] { CHECK ( searchCondition ) | { PRIMARY KEY ( Simple-column-Name [ , Simple-column-Name ]* ) | UNIQUE ( Simple-column-Name [ , Simple-column-Name ]* ) | FOREIGN KEY ( Simple-column-Name [ , Simple-column-Name ]* ) REFERENCES clause } }
References specification
REFERENCES table-Name [ ( Simple-column-Name [ , Simple-column-Name ]* ) ] [ ON DELETE {NO ACTION | RESTRICT | CASCADE | SET NULL}] [ ON UPDATE {NO ACTION | RESTRICT }] | [ ON UPDATE {NO ACTION | RESTRICT }] [ ON DELETE {NO ACTION | RESTRICT | CASCADE | SET NULL}]
searchCondition
A searchCondition is any Boolean expression that meets the requirements specified in Requirements for search condition .
If a constraint-Name is not specified, Derby generates a unique constraint name (for either column or table constraints).
DECLARE GLOBAL TEMPORARY TABLE statement
The DECLARE GLOBAL TEMPORARY TABLE statement defines a temporary table for the current connection. These tables do not reside in the system catalogs and are not persistent. Temporary tables exist only during the connection that declared them and cannot be referenced outside of that connection. When the connection closes, the rows of the table are deleted, and the in-memory description of the temporary table is dropped.
Temporary tables are useful when:
 
the table structure is not known before using an application.
 
other users do not need the same table structure.
 
data in the temporary table is needed while using the application.
 
the table can be declared and dropped without holding the locks on the system catalog.
Syntax
DECLARE GLOBAL TEMPORARY TABLE table-Name { column-definition [ , column-definition ] * } [ ON COMMIT {DELETE | PRESERVE} ROWS ] NOT LOGGED [ON ROLLBACK DELETE ROWS]
table-Name
Names the temporary table. If a schema-Name other than SESSION is specified, an error will occur (SQLSTATE 428EK). If the schema-Name is not specified, SESSION is assigned. Multiple connections can define declared global temporary tables with the same name because each connection has its own unique table descriptor for it.
Using SESSION as the schema name of a physical table will not cause an error, but is discouraged. The SESSION schema name should be reserved for the temporary table schema.
column-definition
See column-definition for CREATE TABLE for more information on column-definition. DECLARE GLOBAL TEMPORARY TABLE does not allow generated-column-spec in the column-definition.
Data-type
Supported data-types are:
 
BIGINT
 
CHAR
 
DATE
 
DECIMAL
 
DOUBLE PRECISION
 
FLOAT
 
INTEGER
 
NUMERIC
 
REAL
 
SMALLINT
 
TIME
 
TIMESTAMP
 
VARCHAR
ON COMMIT
Specifies the action taken on the global temporary table when a COMMIT operation is performed.
DELETE ROWS
All rows of the table will be deleted if no hold-able cursor is open on the table. This is the default value for ON COMMIT. If you specify ON ROLLBACK DELETE ROWS, this will delete all the rows in the table only if the temporary table was used. ON COMMIT DELETE ROWS will delete the rows in the table even if the table was not used (if the table does not have hold-able cursors open on it).
PRESERVE ROWS
The rows of the table will be preserved.
NOT LOGGED
Specifies the action taken on the global temporary table when a rollback operation is performed. When a ROLLBACK (or ROLLBACK TO SAVEPOINT) operation is performed, if the table was created in the unit of work (or savepoint), the table will be dropped. If the table was dropped in the unit of work (or savepoint), the table will be restored with no rows.
ON ROLLBACK DELETE ROWS
This is the default value for NOT LOGGED. NOT LOGGED [ON ROLLBACK DELETE ROWS ]] specifies the action that is to be taken on the global temporary table when a ROLLBACK or (ROLLBACK TO SAVEPOINT) operation is performed. If the table data has been changed, all the rows will be deleted.
Examples
set schema myapp; create table t1(c11 int, c12 date); declare global temporary table SESSION.t1(c11 int) not logged; -- The SESSION qualification is redundant here because temporary -- tables can only exist in the SESSION schema. declare global temporary table t2(c21 int) not logged; -- The temporary table is not qualified here with SESSION because temporary -- tables can only exist in the SESSION schema. insert into SESSION.t1 values (1); -- SESSION qualification is mandatory here if you want to use -- the temporary table, because the current schema is "myapp." select * from t1; -- This select statement is referencing the "myapp.t1" physical -- table since the table was not qualified by SESSION.
Note that temporary tables can only be declared in the SESSION schema. You should never declare a physical schema with the SESSION name.
The following is a list of DB2 UDB DECLARE GLOBAL TEMPORARY TABLE functions that are not supported by Derby:
 
IDENTITY column-options
 
IDENTITY attribute in copy-options
 
AS (fullselect) DEFINITION ONLY
 
NOT LOGGED ON ROLLBACK PRESERVE ROWS
 
IN tablespace-name
 
PARTITIONING KEY
 
WITH REPLACE
Restrictions on Declared Global Temporary Tables
Temporary tables cannot be specified in the following statements:
 
ALTER TABLE
 
CREATE SYNONYM
 
CREATE TRIGGER
 
CREATE VIEW
 
LOCK
 
RENAME
Temporary tables cannot be specified in referential constraints.
There is no check constraints support for columns.
The following data types cannot be used with Declared Global Temporary Tables:
 
BLOB
 
CLOB
 
LONG VARCHAR
Temporary tables cannot be referenced in a triggered-SQL-statement.
If a statement performing an insert, update, or delete to the temporary table encounters an error, all the rows of the table are deleted.
Restrictions Specific to Derby
Derby does not support the following on temporary tables:
 
index support
 
synonyms, triggers and views on SESSION schema tables (including physical tables and temporary tables)
 
LOCK TABLE
 
constraints and primary keys
 
generated-column-spec
 
importing into temporary tables
Any statements referencing SESSION schema tables and views will not be cached.
DELETE statement
Syntax
{ DELETE FROM table-Name [ WHERE clause ] | DELETE FROM table-Name WHERE CURRENT OF }
The first syntactical form, called a searched delete, removes all rows identified by the table name and WHERE clause.
The second syntactical form, called a positioned delete, deletes the current row of an open, updatable cursor. If there is no current row or if it no longer satisfies the cursor's query, an exception is raised. For more information about updatable cursors, see SELECT statement .
Examples
DELETE FROM SAMP.IN_TRAY; stmt.executeUpdate("DELETE FROM SAMP.IN_TRAY WHERE CURRENT OF " + resultSet.getCursorName())
A searched delete statement depends on the table being updated, all of its conglomerates (units of storage such as heaps or indexes), and any other table named in the WHERE clause. A CREATE or DROP INDEX statement for the target table of a prepared searched delete statement invalidates the prepared searched delete statement.
The positioned delete statement depends on the cursor and any tables the cursor references. You can compile a positioned delete even if the cursor has not been opened yet. However, removing the open cursor with the JDBC close method invalidates the positioned delete.
A CREATE or DROP INDEX statement for the target table of a prepared positioned delete invalidates the prepared positioned delete statement.
FOR UPDATE clause
The FOR UPDATE clause is an optional part of a SELECT statement . The FOR UPDATE clause specifies whether the ResultSet of a simple SELECT statement that meets the requirements for a cursor is updatable or not. For more information about updatability, see Requirements for Updatable Cursors .
Syntax
FOR { READ ONLY | FETCH ONLY | UPDATE [ OF Simple-column-Name [ , Simple-column-Name ]* ] }
Simple-column-Name refers to the names visible for the table specified in the FROM clause of the underlying query.
Cursors are read-only by default. For a cursor to be updatable, you must specify FOR UPDATE.
The optimizer is able to use an index even if the column in the index is being updated. For more information about how indexes affect cursors, see Tuning Derby .
SELECT RECEIVED, SOURCE, SUBJECT, NOTE_TEXT FROM SAMP.IN_TRAY FOR UPDATE;
FROM clause
The FROM clause is a mandatory clause in a SelectExpression . It specifies the tables ( TableExpression ) from which the other clauses of the query can access columns for use in expressions.
Syntax
FROM TableExpression [ , TableExpression ] *
SELECT Cities.city_id FROM Cities WHERE city_id < 5 -- other types of TableExpressions SELECT TABLENAME, ISINDEX FROM SYS.SYSTABLES T, SYS.SYSCONGLOMERATES C WHERE T.TABLEID = C.TABLEID ORDER BY TABLENAME, ISINDEX; -- force the join order SELECT * FROM Flights, FlightAvailability WHERE FlightAvailability.flight_id = Flights.flight_id AND FlightAvailability.segment_number = Flights.segment_number AND Flights.flight_id < 'AA1115' -- a TableExpression can be a joinOperation. Therefore -- you can have multiple join operations in a FROM clause SELECT COUNTRIES.COUNTRY, CITIES.CITY_NAME, FLIGHTS.DEST_AIRPORT FROM COUNTRIES LEFT OUTER JOIN CITIES ON COUNTRIES.COUNTRY_ISO_CODE = CITIES.COUNTRY_ISO_CODE LEFT OUTER JOIN FLIGHTS ON Cities.AIRPORT = FLIGHTS.DEST_AIRPORT;
GROUP BY clause
A GROUP BY clause, part of a SelectExpression , groups a result into subsets that have matching values for one or more columns. In each group, no two rows have the same value for the grouping column or columns. NULLs are considered equivalent for grouping purposes.
You typically use a GROUP BY clause in conjunction with an aggregate expression.
Syntax
GROUP BY column-Name [ , column-Name ] *
column-Name must be a column from the current scope of the query; there can be no columns from a query block outside the current scope. For example, if a GROUP BY clause is in a subquery, it cannot refer to columns in the outer query.
SelectItems in the SelectExpression with a GROUP BY clause must contain only aggregates or grouping columns.
-- find the average flying_times of flights grouped by -- airport SELECT AVG (flying_time), orig_airport FROM Flights GROUP BY orig_airport SELECT MAX(city), region FROM Cities, Countries WHERE Cities.country_ISO_code = Countries.country_ISO_code GROUP BY region -- group by an a smallint SELECT ID, AVG(SALARY) FROM SAMP.STAFF GROUP BY ID -- Get the AVGSALARY and EMPCOUNT columns, and the DEPTNO column using the AS clause -- And group by the WORKDEPT column using the correlation name OTHERS SELECT OTHERS.WORKDEPT AS DEPTNO, AVG(OTHERS.SALARY) AS AVGSALARY, COUNT(*) AS EMPCOUNT FROM SAMP.EMPLOYEE OTHERS GROUP BY OTHERS.WORKDEPT;
HAVING clause
A HAVING clause restricts the results of a GROUP BY in a SelectExpression . The HAVING clause is applied to each group of the grouped table, much as a WHERE clause is applied to a select list. If there is no GROUP BY clause, the HAVING clause is applied to the entire result as a single group. The SELECT clause cannot refer directly to any column that does not have a GROUP BY clause. It can, however, refer to constants, aggregates, and special registers.
Syntax
HAVING searchCondition
The searchCondition, which is a specialized booleanExpression, can contain only grouping columns (see GROUP BY clause ), columns that are part of aggregate expressions, and columns that are part of a subquery. For example, the following query is illegal, because the column SALARY is not a grouping column, it does not appear within an aggregate, and it is not within a subquery:
-- SELECT COUNT(*) -- FROM SAMP.STAFF -- GROUP BY ID -- HAVING SALARY > 15000;
Aggregates in the HAVING clause do not need to appear in the SELECT list. If the HAVING clause contains a subquery, the subquery can refer to the outer query block if and only if it refers to a grouping column.
-- Find the total number of economy seats taken on a flight, -- grouped by airline, -- only when the group has at least 2 records. SELECT SUM(ECONOMY_SEATS_TAKEN), AIRLINE_FULL FROM FLIGHTAVAILABILITY, AIRLINES WHERE SUBSTR(FLIGHTAVAILABILITY.FLIGHT_ID, 1, 2) = AIRLINE GROUP BY AIRLINE_FULL HAVING COUNT(*) > 1
INNER JOIN
An INNER JOIN is a JOIN operations that allows you to specify an explicit join clause.
Syntax
TableExpression [ INNER ] JOIN TableExpression { ON booleanExpression }
You can specify the join clause by specifying ON with a boolean expression.
The scope of expressions in the ON clause includes the current tables and any tables in outer query blocks to the current SELECT. In the following example, the ON clause refers to the current tables:
SELECT * FROM SAMP.EMPLOYEE INNER JOIN SAMP.STAFF ON EMPLOYEE.SALARY < STAFF.SALARY;
The ON clause can reference tables not being joined and does not have to reference either of the tables being joined (though typically it does).
-- Join the EMP_ACT and EMPLOYEE tables -- select all the columns from the EMP_ACT table and -- add the employee's surname (LASTNAME) from the EMPLOYEE table -- to each row of the result SELECT SAMP.EMP_ACT.*, LASTNAME FROM SAMP.EMP_ACT JOIN SAMP.EMPLOYEE ON EMP_ACT.EMPNO = EMPLOYEE.EMPNO; -- Join the EMPLOYEE and DEPARTMENT tables, -- select the employee number (EMPNO), -- employee surname (LASTNAME), -- department number (WORKDEPT in the EMPLOYEE table and DEPTNO in the -- DEPARTMENT table) -- and department name (DEPTNAME) -- of all employees who were born (BIRTHDATE) earlier than 1930. SELECT EMPNO, LASTNAME, WORKDEPT, DEPTNAME FROM SAMP.EMPLOYEE JOIN SAMP.DEPARTMENT ON WORKDEPT = DEPTNO AND YEAR(BIRTHDATE) < 1930; -- Another example of "generating" new data values, -- using a query which selects from a VALUES clause (which is an -- alternate form of a fullselect). -- This query shows how a table can be derived called "X" -- having 2 columns "R1" and "R2" and 1 row of data SELECT * FROM (VALUES (3, 4), (1, 5), (2, 6)) AS VALUESTABLE1(C1, C2) JOIN (VALUES (3, 2), (1, 2), (0, 3)) AS VALUESTABLE2(c1, c2) ON VALUESTABLE1.c1 = VALUESTABLE2.c1; This results in: C1 |C2 |C1 |2 ----------------------------------------------- 3 |4 |3 |2 1 |5 |1 |2 -- List every department with the employee number and -- last name of the manager SELECT DEPTNO, DEPTNAME, EMPNO, LASTNAME FROM DEPARTMENT INNER JOIN EMPLOYEE ON MGRNO = EMPNO; -- List every employee number and last name -- with the employee number and last name of their manager SELECT E.EMPNO, E.LASTNAME, M.EMPNO, M.LASTNAME FROM EMPLOYEE E INNER JOIN DEPARTMENT INNER JOIN EMPLOYEE M ON MGRNO = M.EMPNO ON E.WORKDEPT = DEPTNO;
INSERT statement
An INSERT statement creates a row or rows and stores them in the named table. The number of values assigned in an INSERT statement must be the same as the number of specified or implied columns.
Syntax
INSERT INTO table-Name [ ( Simple-column-Name [ , Simple-column-Name ]* ) ] Query
Query can be:
 
 
a VALUES list
 
a multiple-row VALUES expression
Single-row and multiple-row lists can include the keyword DEFAULT. Specifying DEFAULT for a column inserts the column's default value into the column. Another way to insert the default value into the column is to omit the column from the column list and only insert values into other columns in the table. For more information see VALUES Expression .
 
UNION expressions
For more information about Query, see Query .
INSERT INTO COUNTRIES VALUES ('Taiwan', 'TW', 'Asia'); -- Insert a new department into the DEPARTMENT table, -- but do not assign a manager to the new department INSERT INTO DEPARTMENT (DEPTNO, DEPTNAME, ADMRDEPT) VALUES ('E31', 'ARCHITECTURE', 'E01'); -- Insert two new departments using one statement -- into the DEPARTMENT table as in the previous example, -- but do not assign a manager to the new department. INSERT INTO DEPARTMENT (DEPTNO, DEPTNAME, ADMRDEPT) VALUES ('B11', 'PURCHASING', 'B01'), ('E41', 'DATABASE ADMINISTRATION', 'E01'); -- Create a temporary table MA_EMP_ACT with the -- same columns as the EMP_ACT table. -- Load MA_EMP_ACT with the rows from the EMP_ACT -- table with a project number (PROJNO) -- starting with the letters 'MA'. CREATE TABLE MA_EMP_ACT ( EMPNO CHAR(6) NOT NULL, PROJNO CHAR(6) NOT NULL, ACTNO SMALLINT NOT NULL, EMPTIME DEC(5,2), EMSTDATE DATE, EMENDATE DATE ); INSERT INTO MA_EMP_ACT SELECT * FROM EMP_ACT WHERE SUBSTR(PROJNO, 1, 2) = 'MA'; -- Insert the DEFAULT value for the LOCATION column INSERT INTO DEPARTMENT VALUES ('E31', 'ARCHITECTURE', '00390', 'E01', DEFAULT);
Statement dependency system
The INSERT statement depends on the table being inserted into, all of the conglomerates (units of storage such as heaps or indexes) for that table, and any other table named in the query. Any statement that creates or drops an index or a constraint for the target table of a prepared INSERT statement invalidates the prepared INSERT statement.
JOIN operation
The JOIN operations, which are among the possible TableExpression s in a FROM clause , perform joins between two tables. (You can also perform a join between two tables using an explicit equality test in a WHERE clause, such as "WHERE t1.col1 = t2.col2".)
Syntax
JOIN Operation
The JOIN operations are:
 
INNER JOIN
Specifies a join between two tables with an explicit join clause. See INNER JOIN .
 
LEFT OUTER JOIN
Specifies a join between two tables with an explicit join clause, preserving unmatched rows from the first table. See LEFT OUTER JOIN .
 
RIGHT OUTER JOIN
Specifies a join between two tables with an explicit join clause, preserving unmatched rows from the second table. See RIGHT OUTER JOIN .
In all cases, you can specify additional restrictions on one or both of the tables being joined in outer join clauses or in the WHERE clause .
JOIN expressions and query optimization
For information on which types of joins are optimized, see Tuning Derby .
LEFT OUTER JOIN
A LEFT OUTER JOIN is one of the JOIN operations s that allow you to specify a join clause. It preserves the unmatched rows from the first (left) table, joining them with a NULL row in the shape of the second (right) table.
Syntax
TableExpression LEFT [ OUTER ] JOIN TableExpression { ON booleanExpression }
The scope of expressions in either the ON clause includes the current tables and any tables in query blocks outer to the current SELECT. The ON clause can reference tables not being joined and does not have to reference either of the tables being joined (though typically it does).
Example 1
--match cities to countries in Asia SELECT CITIES.COUNTRY, CITIES.CITY_NAME, REGION FROM Countries LEFT OUTER JOIN Cities ON CITIES.COUNTRY_ISO_CODE = COUNTRIES.COUNTRY_ISO_CODE WHERE REGION = 'Asia'; -- use the synonymous syntax, LEFT JOIN, to achieve exactly -- the same results as in the example above SELECT COUNTRIES.COUNTRY, CITIES.CITY_NAME,REGION FROM COUNTRIES LEFT JOIN CITIES ON CITIES.COUNTRY_ISO_CODE = COUNTRIES.COUNTRY_ISO_CODE WHERE REGION = 'Asia';
Example 2
-- Join the EMPLOYEE and DEPARTMENT tables, -- select the employee number (EMPNO), -- employee surname (LASTNAME), -- department number (WORKDEPT in the EMPLOYEE table -- and DEPTNO in the DEPARTMENT table) -- and department name (DEPTNAME) -- of all employees who were born (BIRTHDATE) earlier than 1930 SELECT EMPNO, LASTNAME, WORKDEPT, DEPTNAME FROM SAMP.EMPLOYEE LEFT OUTER JOIN SAMP.DEPARTMENT ON WORKDEPT = DEPTNO AND YEAR(BIRTHDATE) < 1930; -- List every department with the employee number and -- last name of the manager, -- including departments without a manager SELECT DEPTNO, DEPTNAME, EMPNO, LASTNAME FROM DEPARTMENT LEFT OUTER JOIN EMPLOYEE ON MGRNO = EMPNO;
LOCK TABLE statement
Allows a user to explicitly acquire a shared or exclusive table lock on the specified table. The table lock lasts until the end of the current transaction.
Explicitly locking a table is useful for:
 
avoiding the overhead of multiple row locks on a table (in other words, user-initiated lock escalation)
 
avoiding deadlocks
You cannot lock system tables with this statement.
Syntax
LOCK TABLE table-Name IN { SHARE | EXCLUSIVE } MODE
Once a table is locked in either mode, a transaction does not acquire any subsequent row-level locks on a table. Replace line 13 with this: For example, if a transaction locks the entire Flights table in share mode in order to read data, a particular statement might need to lock a particular row in exclusive mode in order to update the row. However, the previous table-level lock on Hotels forces the exclusive lock to be table-level as well.
If the specified lock cannot be acquired because another connection already holds a lock on the table, a statement-level exception is raised (SQLState X0X02) after the deadlock timeout period.
-- lock the entire table in share mode to avoid -- a large number of row locks LOCK TABLE Flights IN SHARE MODE; SELECT * FROM Flights WHERE orig_airport > 'OOO'; -- lock the entire table in exclusive mode -- for a transaction that will update many rows, -- but where no single statement will update enough rows -- acquire an exclusive table lock on the table. -- In a row-level locking system, that transaction would -- require a large number of locks or might deadlock. LOCK TABLE HotelAvailability IN EXCLUSIVE MODE; UPDATE HotelAvailability SET rooms_taken = (rooms_taken + 2) WHERE hotel_id = 194 AND booking_date = DATE('1998-04-10'); UPDATE HotelAvailability SET rooms_taken = (rooms_taken + 2) WHERE hotel_id = 194 AND booking_date = DATE('1998-04-11'); UPDATE HotelAvailability SET rooms_taken = (rooms_taken + 2) WHERE hotel_id = 194 AND booking_date = DATE('1998-04-12'); UPDATE HotelAvailability SET rooms_taken = (rooms_taken + 2) WHERE hotel_id = 194 AND booking_date = DATE('1998-04-12'); -- if a transaction needs to look at a table before -- updating it, acquire an exclusive lock before -- selecting to avoid deadlocks LOCK TABLE People IN EXCLUSIVE MODE; SELECT MAX(person_id) + 1 FROM PEOPLE; -- INSERT INTO PEOPLE . . .
ORDER BY clause
The ORDER BY clause is an optional element of a SELECT statement . An ORDER BY clause allows you to specify the order in which rows appear in the ResultSet.
Syntax
ORDER BY { column-Name | ColumnPosition } [ ASC | DESC ] [ , column-Name | ColumnPosition [ ASC | DESC ] ] *
ColumnPosition is an integer that identifies the number of the column in the SelectItem in the underlying Query of the SELECT statement . ColumnPosition must be greater than 0 and not greater than the number of columns in the result table. In other words, if you want to order by a column, that column must be in the select list.
column-Name refers to the names visible from the SelectItems in the underlying query of the SELECT statement . An order by column does not need to be in the select list.
ASC specifies that the results should be returned in ascending order; DESC specifies that the results should be returned in descending order. If the order is not specified, ASC is the default.
An ORDER BY clause prevents a SELECT statement from being an updatable cursor. (For more information, see Requirements for updatable cursors and updatable ResultSets .)
For example, if an INTEGER column contains integers, NULL is considered greater than 1 for purposes of sorting. In other words, NULL values are sorted high.
-- order by the correlation name NATION -- order by the correlation name NATION SELECT CITY_NAME, COUNTRY AS NATION FROM CITIES ORDER BY NATION;
Query
A query creates a virtual table based on existing tables or constants built into tables.
Syntax
{ ( Query ) | Query INTERSECT [ ALL | DISTINCT ] Query | Query EXCEPT [ ALL | DISTINCT ] Query | Query UNION [ ALL | DISTINCT ] Query | SelectExpression | VALUES Expression }
You can arbitrarily put parentheses around queries, or use the parentheses to control the order of evaluation of the INTERSECT, EXCEPT, or UNION operations. These operations are evaluated from left to right when no parentheses are present, with the exception of INTERSECT operations, which would be evaluated before any UNION or EXCEPT operations.
Duplicates in UNION, INTERSECT, and EXCEPT ALL results
The ALL and DISTINCT keywords determine whether duplicates are eliminated from the result of the operation. If you specify the DISTINCT keyword, then the result will have no duplicate rows. If you specify the ALL keyword, then there may be duplicates in the result, depending on whether there were duplicates in the input. DISTINCT is the default, so if you don't specify ALL or DISTINCT, the duplicates will be eliminated. For example, UNION builds an intermediate ResultSet with all of the rows from both queries and eliminates the duplicate rows before returning the remaining rows. UNION ALL returns all rows from both queries as the result.
Depending on which operation is specified, if the number of copies of a row in the left table is L and the number of copies of that row in the right table is R, then the number of duplicates of that particular row that the output table contains (assuming the ALL keyword is specified) is:
 
UNION: ( L + R ).
 
EXCEPT: the maximum of ( L – R ) and 0 (zero).
 
INTERSECT: the minimum of L and R.
Examples
-- a Select expression SELECT * FROM ORG; -- a subquery SELECT * FROM (SELECT CLASS_CODE FROM CL_SCHED) AS CS; -- a subquery SELECT * FROM (SELECT CLASS_CODE FROM CL_SCHED) AS CS (CLASS_CODE); -- a UNION -- returns all rows from columns DEPTNUMB and MANAGER -- in table ORG -- and (1,2) and (3,4) -- DEPTNUMB and MANAGER are smallint columns SELECT DEPTNUMB, MANAGER FROM ORG UNION ALL VALUES (1,2), (3,4); -- a values expression VALUES (1,2,3); -- List the employee numbers (EMPNO) of all employees in the EMPLOYEE table -- whose department number (WORKDEPT) either begins with 'E' or -- who are assigned to projects in the EMP_ACT table -- whose project number (PROJNO) equals 'MA2100', 'MA2110', or 'MA2112' SELECT EMPNO FROM EMPLOYEE WHERE WORKDEPT LIKE 'E%' UNION SELECT EMPNO FROM EMP_ACT WHERE PROJNO IN('MA2100', 'MA2110', 'MA2112'); -- Make the same query as in the previous example -- and "tag" the rows from the EMPLOYEE table with 'emp' and -- the rows from the EMP_ACT table with 'emp_act'. -- Unlike the result from the previous example, -- this query may return the same EMPNO more than once, -- identifying which table it came from by the associated "tag" SELECT EMPNO, 'emp' FROM EMPLOYEE WHERE WORKDEPT LIKE 'E%' UNION SELECT EMPNO, 'emp_act' FROM EMP_ACT WHERE PROJNO IN('MA2100', 'MA2110', 'MA2112'); -- Make the same query as in the previous example, -- only use UNION ALL so that no duplicate rows are eliminated SELECT EMPNO FROM EMPLOYEE WHERE WORKDEPT LIKE 'E%' UNION ALL SELECT EMPNO FROM EMP_ACT WHERE PROJNO IN('MA2100', 'MA2110', 'MA2112'); -- Make the same query as in the previous example, -- only include an additional two employees currently not in any table and -- tag these rows as "new" SELECT EMPNO, 'emp' FROM EMPLOYEE WHERE WORKDEPT LIKE 'E%' UNION SELECT EMPNO, 'emp_act' FROM EMP_ACT WHERE PROJNO IN('MA2100', 'MA2110', 'MA2112') UNION VALUES ('NEWAAA', 'new'), ('NEWBBB', 'new');
RIGHT OUTER JOIN
A RIGHT OUTER JOIN is one of the JOIN operations s that allow you to specify a JOIN clause. It preserves the unmatched rows from the second (right) table, joining them with a NULL in the shape of the first (left) table. A LEFT OUTER JOIN B is equivalent to B RIGHT OUTER JOIN A, with the columns in a different order.
Syntax
TableExpression RIGHT [ OUTER ] JOIN TableExpression { ON booleanExpression }
The scope of expressions in the ON clause includes the current tables and any tables in query blocks outer to the current SELECT. The ON clause can reference tables not being joined and does not have to reference either of the tables being joined (though typically it does).
Example 1
-- get all countries and corresponding cities, including -- countries without any cities SELECT COUNTRIES.COUNTRY, CITIES.CITY_NAME FROM CITIES RIGHT OUTER JOIN COUNTRIES ON CITIES.COUNTRY_ISO_CODE = COUNTRIES.COUNTRY_ISO_CODE; -- get all countries in Africa and corresponding cities, including -- countries without any cities SELECT COUNTRIES.COUNTRY, CITIES.CITY_NAME FROM CITIES RIGHT OUTER JOIN COUNTRIES ON CITIES.COUNTRY_ISO_CODE = COUNTRIES.COUNTRY_ISO_CODE WHERE Countries.region = 'Africa'; -- use the synonymous syntax, RIGHT JOIN, to achieve exactly -- the same results as in the example above SELECT COUNTRIES.COUNTRY, CITIES.CITY_NAME FROM CITIES RIGHT JOIN COUNTRIES ON CITIES.COUNTRY_ISO_CODE = COUNTRIES.COUNTRY_ISO_CODE WHERE Countries.region = 'Africa';
Example 2
-- a TableExpression can be a joinOperation. Therefore -- you can have multiple join operations in a FROM clause -- List every employee number and last name -- with the employee number and last name of their manager SELECT E.EMPNO, E.LASTNAME, M.EMPNO, M.LASTNAME FROM EMPLOYEE E RIGHT OUTER JOIN DEPARTMENT RIGHT OUTER JOIN EMPLOYEE M ON MGRNO = M.EMPNO ON E.WORKDEPT = DEPTNO;
ScalarSubquery
You can place a ScalarSubquery anywhere an Expression is permitted. A ScalarSubquery turns a SelectExpression result into a scalar value because it returns only a single row and column value.
The query must evaluate to a single row with a single column.
Sometimes also called an expression subquery.
Syntax
( Query )
-- avg always returns a single value, so the subquery is -- a ScalarSubquery SELECT NAME, COMM FROM STAFF WHERE EXISTS (SELECT AVG(BONUS + 800) FROM EMPLOYEE WHERE COMM < 5000 AND EMPLOYEE.LASTNAME = UPPER(STAFF.NAME) ); -- Introduce a way of "generating" new data values, -- using a query which selects from a VALUES clause (which is an alternate form of a fullselect). -- This query shows how a table can be derived called "X" having 2 columns "R1" and "R2" and 1 row of data. SELECT R1,R2 FROM (VALUES('GROUP 1','GROUP 2')) AS X(R1,R2);
SelectExpression
A SelectExpression is the basic SELECT-FROM-WHERE construct used to build a table value based on filtering and projecting values from other tables.
Syntax
SELECT [ DISTINCT | ALL ] SelectItem [ , SelectItem ]* FROM clause [ WHERE clause ] [ GROUP BY clause ] [ HAVING clause ]
SelectItem:
{ * | { table-Name | correlation-Name } .* | Expression [AS Simple-column-Name ] }
The SELECT clause contains a list of expressions and an optional quantifier that is applied to the results of the FROM clause and the WHERE clause . If DISTINCT is specified, only one copy of any row value is included in the result. Nulls are considered duplicates of one another for the purposes of DISTINCT. If no quantifier, or ALL, is specified, no rows are removed from the result in applying the SELECT clause (ALL is the default).
A SelectItem projects one or more result column values for a table result being constructed in a SelectExpression.
The result of the FROM clause is the cross product of the FROM items. The WHERE clause can further qualify this result.
The WHERE clause causes rows to be filtered from the result based on a boolean expression. Only rows for which the expression evaluates to TRUE are returned in the result.
The GROUP BY clause groups rows in the result into subsets that have matching values for one or more columns. GROUP BY clauses are typically used with aggregates.
If there is a GROUP BY clause, the SELECT clause must contain only aggregates or grouping columns. If you want to include a non-grouped column in the SELECT clause, include the column in an aggregate expression. For example:
-- List head count of each department, -- the department number (WORKDEPT), and the average departmental salary (SALARY) -- for all departments in the EMPLOYEE table. -- Arrange the result table in ascending order by average departmental salary. SELECT WORKDEPT, AVG(SALARY) FROM EMPLOYEE GROUP BY WORKDEPT ORDER BY 1;
If there is no GROUP BY clause, but a SelectItem contains an aggregate not in a subquery, the query is implicitly grouped. The entire table is the single group.
The HAVING clause restricts a grouped table, specifying a search condition (much like a WHERE clause) that can refer only to grouping columns or aggregates from the current scope. The HAVING clause is applied to each group of the grouped table. If the HAVING clause evaluates to TRUE, the row is retained for further processing. If the HAVING clause evaluates to FALSE or NULL, the row is discarded. If there is a HAVING clause but no GROUP BY, the table is implicitly grouped into one group for the entire table.
Derby processes a SelectExpression in the following order:
 
FROM clause
 
WHERE clause
 
GROUP BY (or implicit GROUP BY)
 
HAVING clause
 
SELECT clause
The result of a SelectExpression is always a table.
When a query does not have a FROM clause (when you are constructing a value, not getting data out of a table), you use a VALUES statement, not a SelectExpression. For example:
VALUES CURRENT_TIMESTAMP
The * wildcard
* expands to all columns in the tables in the associated FROM clause.
table-Name .* and correlation-Name .* expand to all columns in the identified table. That table must be listed in the associated FROM clause.
Naming columns
You can name a SelectItem column using the AS clause. When the SelectExpression appears in a UNION, INTERSECT, or EXCEPT operator, the names from the first SelectExpression are taken as the names for the columns in the result of the operation. If a column of a SelectItem is not a simple ColumnReference expression or named with an AS clause, it is given a generated unique name.
These column names are useful in several cases:
 
They are made available on the JDBC ResultSetMetaData.
 
They are used as the names of the columns in the resulting table when the SelectExpression is used as a table subquery in a FROM clause.
 
They are used in the ORDER BY clause as the column names available for sorting.
-- this example shows SELECT-FROM-WHERE -- with an ORDER BY clause -- and correlation-Names for the tables SELECT CONSTRAINTNAME, COLUMNNAME FROM SYS.SYSTABLES t, SYS.SYSCOLUMNS col, SYS.SYSCONSTRAINTS cons, SYS.SYSCHECKS checks WHERE t.TABLENAME = 'FLIGHTS' AND t.TABLEID = col. REFERENCEID AND t.TABLEID = cons.TABLEID AND cons.CONSTRAINTID = checks.CONSTRAINTID ORDER BY CONSTRAINTNAME; -- This example shows the use of the DISTINCT clause SELECT DISTINCT ACTNO FROM EMP_ACT; -- This example shows how to rename an expression -- Using the EMPLOYEE table, list the department number (WORKDEPT) and -- maximum departmental salary (SALARY) renamed as BOSS -- for all departments whose maximum salary is less than the -- average salary in all other departments. SELECT WORKDEPT AS DPT, MAX(SALARY) AS BOSS FROM EMPLOYEE EMP_COR GROUP BY WORKDEPT HAVING MAX(SALARY) < (SELECT AVG(SALARY) FROM EMPLOYEE WHERE NOT WORKDEPT = EMP_COR.WORKDEPT) ORDER BY BOSS;
SELECT statement
A SELECT statement consists of a query with an optional ORDER BY clause and an optional FOR UPDATE clause . The SELECT statement is so named because the typical first word of the query construct is SELECT. (Query includes the VALUES expression and UNION, INTERSECT, and EXCEPT expressions as well as SELECT expressions).
The ORDER BY clause guarantees the ordering of the ResultSet. The FOR UPDATE clause makes the result an updatable cursor. The SELECT statement supports the FOR FETCH ONLY clause. The FOR FETCH ONLY clause is synonymous with the FOR READ ONLY clause.
Remember: In order to get an updatable ResultSet, you must include a FOR UPDATE clause with the SELECT clause.
Syntax
Query [ ORDER BY clause ] [ FOR UPDATE clause ] WITH {RR|RS|CS|UR}
You can set the isolation level in a SELECT statement using the WITH {RR|RS|CS|UR} syntax.
-- lists the names of the expression SAL+BONUS+COMM as TOTAL_PAY and -- orders by the new name TOTAL_PAY SELECT FIRSTNME, SALARY+BONUS+COMM AS TOTAL_PAY FROM EMPLOYEE ORDER BY TOTAL_PAY; -- creating an updatable cursor with a FOR UPDATE clause -- to update the start date (PRSTDATE) and the end date (PRENDATE) -- columns in the PROJECT table SELECT PROJNO, PRSTDATE, PRENDATE FROM PROJECT FOR UPDATE OF PRSTDATE, PRENDATE; -- set the isolation level to RR for this statement only SELECT * FROM Flights WHERE flight_id BETWEEN 'AA1111' AND 'AA1112' WITH RR;
A SELECT statement returns a ResultSet. A cursor is a pointer to a specific row in ResultSet. In Java applications, all ResultSets are cursors. A cursor is updatable; that is, you can update or delete rows as you step through the ResultSet if the SELECT statement that generated it and its underlying query meet cursor updatability requirements, as detailed below. You use a FOR UPDATE clause when you want to generate an updatable cursor.
Note: The ORDER BY clause allows you to order the results of the SELECT. Without the ORDER BY clause, the results are returned in random order.
If a SELECT statement meets the requirements listed below, cursors are updatable only if you specify FOR UPDATE in the FOR clause (see FOR UPDATE clause ).
Requirements for updatable cursors and updatable ResultSets
Only simple, single-table SELECT cursors and FORWARD_ONLY ResultSets can be updatable. The SELECT statement for updatable ResultSets has the same syntax as the SELECT statement for updatable cursors. To generate updatable cursors:
 
The SELECT statement must not include an ORDER BY clause.
 
The underlying Query must be a SelectExpression .
 
The SelectExpression in the underlying Query must not include:
 
DISTINCT
 
Aggregates
 
GROUP BY clause
 
HAVING clause
 
The FROM clause in the underlying Query must not have:
 
more than one table in its FROM clause
 
anything other than one table name
 
 
subqueries
There is no SQL language statement to assign a name to a cursor. Instead, you use the JDBC API to assign names to cursors or retrieve system-generated names. For more information, see "Naming or Accessing the Name of a Cursor" in Chapter 5 of the Derby Developer's Guide .
Cursors are read-only by default. For a cursor to be updatable, you must specify FOR UPDATE in the FOR clause (see FOR UPDATE clause ).
Statement dependency system
The SELECT depends on all the tables and views named in the query and the conglomerates (units of storage such as heaps and indexes) chosen for access paths on those tables. CREATE INDEX does not invalidate a prepared SELECT statement. A DROP INDEX statement invalidates a prepared SELECT statement if the index is an access path in the statement. If the SELECT includes views, it also depends on the dictionary objects on which the view itself depends (see CREATE VIEW statement ).
Any prepared UPDATE WHERE CURRENT or DELETE WHERE CURRENT statement against a cursor of a SELECT depends on the SELECT. Removing a SELECT through a java.sql.Statement.close request invalidates the UPDATE WHERE CURRENT or DELETE WHERE CURRENT.
The SELECT depends on all aliases used in the query. Dropping an alias invalidates a prepared SELECT statement if the statement uses the alias.
TableExpression
A TableExpression specifies a table or view in a FROM clause . It is the source from which a SelectExpression selects a result.
A correlation name can be applied to a table in a TableExpression so that its columns can be qualified with that name. If you do not supply a correlation name, the table name qualifies the column name. When you give a table a correlation name, you cannot use the table name to qualify columns. You must use the correlation name when qualifying column names.
No two items in the FROM clause can have the same correlation name, and no correlation name can be the same as an unqualified table name specified in that FROM clause.
In addition, you can give the columns of the table new names in the AS clause. Some situations in which this is useful:
 
When a VALUES expression is used as a TableSubquery , since there is no other way to name the columns of a VALUES expression .
 
When column names would otherwise be the same as those of columns in other tables; renaming them means you don't have to qualify them.
The Query in a TableSubquery appearing in a FromItem can contain multiple columns and return multiple rows. See TableSubquery .
For information about the optimizer overrides you can specify, see Tuning Derby .
Syntax
{ TableOrViewExpression | JOIN operation }
-- SELECT from a Join expression SELECT E.EMPNO, E.LASTNAME, M.EMPNO, M.LASTNAME FROM EMPLOYEE E LEFT OUTER JOIN DEPARTMENT INNER JOIN EMPLOYEE M ON MGRNO = M.EMPNO ON E.WORKDEPT = DEPTNO
TableOrViewExpression
TableSubquery
A TableSubquery is a subquery that returns multiple rows.
Unlike a ScalarSubquery , a TableSubquery is allowed only:
 
 
with EXISTS, IN, or quantified comparisons.
When used as a TableExpression in a FROM clause , it can return multiple columns. When used with EXISTS, it returns multiple columns only if you use * to return the multiple columns.
When used with IN or quantified comparisons, it must return a single column.
Syntax
( Query )
-- a subquery used as a TableExpression in a FROM clause SELECT VirtualFlightTable.flight_ID FROM (SELECT flight_ID, orig_airport, dest_airport FROM Flights WHERE (orig_airport = 'SFO' OR dest_airport = 'SCL') ) AS VirtualFlightTable -- a subquery (values expression) used as a TableExpression -- in a FROM clause SELECT mycol1 FROM (VALUES (1, 2), (3, 4)) AS mytable (mycol1, mycol2) -- a subquery used with EXISTS SELECT * FROM Flights WHERE EXISTS (SELECT * FROM Flights WHERE dest_airport = 'SFO' AND orig_airport = 'GRU') -- a subquery used with IN SELECT flight_id, segment_number FROM Flights WHERE flight_id IN (SELECT flight_ID FROM Flights WHERE orig_airport = 'SFO' OR dest_airport = 'SCL') -- a subquery used with a quantified comparison SELECT NAME, COMM FROM STAFF WHERE COMM > (SELECT AVG(BONUS + 800) FROM EMPLOYEE WHERE COMM < 5000);
UPDATE statement
An UPDATE statement sets the value in a column.
You can update the current row of an open, updatable cursor. If there is no current row, or if the current row no longer satisfies the cursor's query, an exception is raised.
Syntax
The first syntactical form is called a searched update. The second syntactical form is called a positioned update.
For searched updates, you update all rows of the table for which the WHERE clause evaluates to TRUE.
For positioned updates, you can update only columns that were included in the FOR UPDATE clause of the SELECT statement that created the cursor. If the SELECT statement did not include a FOR UPDATE clause, the cursor is read-only and cannot be used to update.
Specifying DEFAULT for the update value sets the value of the column to the default defined for that table.
-- All the employees except the manager of department (WORKDEPT) 'E21' have been temporarily reassigned. -- Indicate this by changing their job (JOB) to NULL and their pay -- (SALARY, BONUS, COMM) values to zero in the EMPLOYEE table. UPDATE EMPLOYEE SET JOB=NULL, SALARY=0, BONUS=0, COMM=0 WHERE WORKDEPT = 'E21' AND JOB <> 'MANAGER' -- PROMOTE the job (JOB) of certain employees to MANAGER UPDATE EMPLOYEE SET JOB = 'MANAGER' WHERE CURRENT OF CURS1; -- Increase the project staffing (PRSTAFF) by 1.5 for all projects stmt.executeUpdate("UPDATE PROJECT SET PRSTAFF = " "PRSTAFF + 1.5" + "WHERE CURRENT OF" + ResultSet.getCursorName()); -- Change the job (JOB) of employee number (EMPNO) '000290' in the EMPLOYEE table -- to its DEFAULT value which is NULL UPDATE EMPLOYEE SET JOB = DEFAULT WHERE EMPNO = '000290';
Statement dependency system
A searched update statement depends on the table being updated, all of its conglomerates (units of storage such as heaps or indexes), all of its constraints, and any other table named in the WHERE clause or SET expressions. A CREATE or DROP INDEX statement or an ALTER TABLE statement for the target table of a prepared searched update statement invalidates the prepared searched update statement.
The positioned update statement depends on the cursor and any tables the cursor references. You can compile a positioned update even if the cursor has not been opened yet. However, removing the open cursor with the JDBC close method invalidates the positioned update.
A CREATE or DROP INDEX statement or an ALTER TABLE statement for the target table of a prepared positioned update invalidates the prepared positioned update statement.
Dropping an alias invalidates a prepared update statement if the latter statement uses the alias.
Dropping or adding triggers on the target table of the update invalidates the update statement.
Value
Expression | DEFAULT
VALUES expression
The VALUES expression allows construction of a row or a table from other values. You use a VALUES statement when you do not have a FROM clause. This construct can be used in all the places where a query can, and thus can be used as a statement that returns a ResultSet, within expressions and statements wherever subqueries are permitted, and as the source of values for an INSERT statement.
Syntax
{ VALUES ( Value {, Value }* ) [ , ( Value {, Value }* ) ]* | VALUES Value [ , Value ]* | }
The first form constructs multi-column rows. The second form constructs single-column rows, each expression being the value of the column of the row.
The DEFAULT keyword is allowed only if the VALUES expression is in an INSERT statement. Specifying DEFAULT for a column inserts the column's default value into the column. Another way to insert the default value into the column is to omit the column from the column list and only insert values into other columns in the table.
-- 3 rows of 1 column VALUES (1),(2),(3); -- 3 rows of 1 column VALUES 1, 2, 3; -- 1 row of 3 columns VALUES (1, 2, 3); -- 3 rows of 2 columns VALUES (1,21),(2,22),(3,23); -- constructing a derived table VALUES ('orange', 'orange'), ('apple', 'red'), ('banana', 'yellow') -- Insert two new departments using one statement into the DEPARTMENT table, -- but do not assign a manager to the new department. INSERT INTO DEPARTMENT (DEPTNO, DEPTNAME, ADMRDEPT) VALUES ('B11', 'PURCHASING', 'B01'), ('E41', 'DATABASE ADMINISTRATION', 'E01') -- insert a row with a DEFAULT value for the MAJPROJ column INSERT INTO PROJECT (PROJNO, PROJNAME, DEPTNO, RESPEMP, PRSTDATE, MAJPROJ) VALUES ('PL2101', 'ENSURE COMPAT PLAN', 'B01', '000020', CURRENT_DATE, DEFAULT); -- using a built-in function VALUES CURRENT_DATE -- getting the value of an arbitrary expression VALUES (3*29, 26.0E0/3) -- getting a value returned by a built-in function values char(1)
Value
Expression | DEFAULT
WHERE clause
A WHERE clause is an optional part of a SelectExpression , DELETE statement , or UPDATE statement . The WHERE clause lets you select rows based on a boolean expression. Only rows for which the expression evaluates to TRUE are returned in the result, or, in the case of a DELETE statement, deleted, or, in the case of an UPDATE statement, updated.
Syntax
Boolean expressions are allowed in the WHERE clause. Most of the general expressions listed in Table of Expressions , can result in a boolean value.
In addition, there are the more common boolean expressions. Specific boolean operators listed in Table 10, take one or more operands; the expressions return a boolean value.
-- find the flights where no business-class seats have -- been booked SELECT * FROM FlightAvailability WHERE business_seats_taken IS NULL OR business_seats_taken = 0 -- Join the EMP_ACT and EMPLOYEE tables -- select all the columns from the EMP_ACT table and -- add the employee's surname (LASTNAME) from the EMPLOYEE table -- to each row of the result. SELECT SAMP.EMP_ACT.*, LASTNAME FROM SAMP.EMP_ACT, SAMP.EMPLOYEE WHERE EMP_ACT.EMPNO = EMPLOYEE.EMPNO; -- Determine the employee number and salary of sales representatives -- along with the average salary and head count of their departments. -- This query must first create a new-column-name specified in the AS clause -- which is outside the fullselect (DINFO) -- in order to get the AVGSALARY and EMPCOUNT columns, -- as well as the DEPTNO column that is used in the WHERE clause SELECT THIS_EMP.EMPNO, THIS_EMP.SALARY, DINFO.AVGSALARY, DINFO.EMPCOUNT FROM EMPLOYEE THIS_EMP, (SELECT OTHERS.WORKDEPT AS DEPTNO, AVG(OTHERS.SALARY) AS AVGSALARY, COUNT(*) AS EMPCOUNT FROM EMPLOYEE OTHERS GROUP BY OTHERS.WORKDEPT )AS DINFO WHERE THIS_EMP.JOB = 'SALESREP' AND THIS_EMP.WORKDEPT = DINFO.DEPTNO;
WHERE CURRENT OF clause
The WHERE CURRENT OF clause is a clause in some UPDATE and DELETE statements. It allows you to perform positioned updates and deletes on updatable cursors. For more information about updatable cursors, see SELECT statement .
Syntax
WHERE CURRENT OF cursor-Name
Statement s = conn.createStatement(); s.setCursorName("AirlinesResults"); ResultSet rs = conn.executeQuery( "SELECT Airline, basic_rate " + "FROM Airlines FOR UPDATE OF basic_rate"); Statement s2 = conn.createStatement(); s2.executeUpdate("UPDATE Airlines SET basic_rate = basic_rate " + "+ .25 WHERE CURRENT OF AirlinesResults");
Built-in functions
A built-in function is an expression in which an SQL keyword or special operator executes some operation. Built-in functions use keywords or special built-in operators. Built-ins are SQL92Identifiers and are case-insensitive. Note that escaped functions like TIMESTAMPADD and TIMESTAMPDIFF are only accessible using the JDBC escape function syntax, and can be found in JDBC escape syntax .
Standard built-in functions
 
 
 
 
 
 
 
 
 
 
 
 
 
DAY
 
 
 
 
 
 
 
 
 
 
MOD
 
 
 
 
 
 
 
 
 
 
 
 
 
Aggregates (set functions)
This section describes aggregates (also described as set functions in ANSI SQL-92 and as column functions in some database literature). They provide a means of evaluating an expression over a set of rows. Whereas the other built-in functions operate on a single expression, aggregates operate on a set of values and reduce them to a single scalar value. Built-in aggregates can calculate the minimum, maximum, sum, count, and average of an expression over a set of values as well as count rows. You can also create your own aggregates to perform other set functions such as calculating the standard deviation.
The built-in aggregates can operate on the data types shown in Permitted Data Types for Built-in Aggregates .
Table1. Permitted Data Types for Built-in Aggregates
 '
All Types
Numeric Built-in Data Types
COUNT
X
X
MIN
 '
X
MAX
 '
X
AVG
 '
X
SUM
 '
X
Aggregates are permitted only in the following:
 
A SelectItem in a SelectExpression .
 
 
An ORDER BY clause (using an alias name) if the aggregate appears in the result of the relevant query block. That is, an alias for an aggregate is permitted in an ORDER BY clause if and only if the aggregate appears in a SelectItem in a SelectExpression .
All expressions in SelectItems in the SelectExpression must be either aggregates or grouped columns (see GROUP BY clause ). (The same is true if there is a HAVING clause without a GROUP BY clause.) This is because the ResultSet of a SelectExpression must be either a scalar (single value) or a vector (multiple values), but not a mixture of both. (Aggregates evaluate to a scalar value, and the reference to a column can evaluate to a vector.) For example, the following query mixes scalar and vector values and thus is not valid:
-- not valid SELECT MIN(flying_time), flight_id FROM Flights
Aggregates are not allowed on outer references (correlations). This means that if a subquery contains an aggregate, that aggregate cannot evaluate an expression that includes a reference to a column in the outer query block. For example, the following query is not valid because SUM operates on a column from the outer query:
SELECT c1 FROM t1 GROUP BY c1 HAVING c2 > (SELECT t2.x FROM t2 WHERE t2.y = SUM(t1.c3))
A cursor declared on a ResultSet that includes an aggregate in the outer query block is not updatable.
This section includes the following aggregates:
 
AVG
 
 
MAX
 
MIN
 
SUM
ABS or ABSVAL
ABS or ABSVAL returns the absolute value of a numeric expression. The return type is the type of parameter. All built-in numeric types are supported ( DECIMAL , DOUBLE PRECISION , FLOAT , INTEGER , BIGINT , NUMERIC , REAL , and SMALLINT ).
Syntax
ABS(NumericExpression)
-- returns 3 VALUES ABS(-3)
AVG
AVG is an aggregate function that evaluates the average of an expression over a set of rows (see Aggregates (set functions) ). AVG is allowed only on expressions that evaluate to numeric data types.
Syntax
AVG ( [ DISTINCT | ALL ] Expression )
The DISTINCT qualifier eliminates duplicates. The ALL qualifier retains duplicates. ALL is the default value if neither ALL nor DISTINCT is specified. For example, if a column contains the values 1.0, 1.0, 1.0, 1.0, and 2.0, AVG(col) returns a smaller value than AVG(DISTINCT col).
Only one DISTINCT aggregate expression per SelectExpression is allowed. For example, the following query is not valid:
SELECT AVG (DISTINCT flying_time), SUM (DISTINCT miles) FROM Flights
The expression can contain multiple column references or expressions, but it cannot contain another aggregate or subquery. It must evaluate to an SQL-92 numeric data type. You can therefore call methods that evaluate to SQL-92 data types. If an expression evaluates to NULL, the aggregate skips that value.
The resulting data type is the same as the expression on which it operates (it will never overflow). The following query, for example, returns the INTEGER 1, which might not be what you would expect:
SELECT AVG(c1) FROM (VALUES (1), (1), (1), (1), (2)) AS myTable (c1)
CAST the expression to another data type if you want more precision:
SELECT AVG(CAST (c1 AS DOUBLE PRECISION)) FROM (VALUES (1), (1), (1), (1), (2)) AS myTable (c1)
BIGINT
The BIGINT function returns a 64-bit integer representation of a number or character string in the form of an integer constant.
Syntax
BIGINT (CharacterExpression | NumericExpression )
CharacterExpression
An expression that returns a character string value of length not greater than the maximum length of a character constant. Leading and trailing blanks are eliminated and the resulting string must conform to the rules for forming an SQL integer constant. The character string cannot be a long string. If the argument is a CharacterExpression, the result is the same number that would occur if the corresponding integer constant were assigned to a big integer column or variable.
NumericExpression
An expression that returns a value of any built-in numeric data type. If the argument is a NumericExpression, the result is the same number that would occur if the argument were assigned to a big integer column or variable. If the whole part of the argument is not within the range of integers, an error occurs. The decimal part of the argument is truncated if present.
The result of the function is a big integer. If the argument can be null, the result can be null; if the argument is null, the result is the null value.
Using the EMPLOYEE table, select the EMPNO column in big integer form for further processing in the application:
SELECT BIGINT (EMPNO) FROM EMPLOYEE
CAST
CAST converts a value from one data type to another and provides a data type to a dynamic parameter (?) or a NULL value.
CAST expressions are permitted anywhere expressions are permitted.
Syntax
CAST ( [ Expression | NULL | ? ] AS Datatype)
The data type to which you are casting an expression is the target type. The data type of the expression from which you are casting is the source type.
CAST conversions among SQL-92 data types
The following table shows valid explicit conversions between source types and target types for SQL data types.
Table1. Explicit conversions between source types and target types for SQL data types
This table shows which explicit conversions between data types are valid. The first column on the table lists the source types, while the first row lists the target types. A "Y" indicates that the source to the target is a valid conversion.
Types
S M A L L I N T
I N T E G E R
B I G I N T
D E C I M A L
R E A L
D O U B L E
F L O A T
C H A R
V A R C H A R
L O N G V A R C H A R
C H A R F O R B I T D A T A
V A R C H A R F O R B I T D A T A
L O N G V A R C H A R F O R B I T D A T A
C L O B
B L O B
D A T E
T I M E
T I M E S T A M P
SMALLINT
Y
Y
Y
Y
Y
Y
Y
Y
-
-
-
-
-
-
-
-
-
-
INTEGER
Y
Y
Y
Y
Y
Y
Y
Y
-
-
-
-
-
-
-
-
-
-
BIGINT
Y
Y
Y
Y
Y
Y
Y
Y
-
-
-
-
-
-
-
-
-
-
DECIMAL
Y
Y
Y
Y
Y
Y
Y
Y
-
-
-
-
-
-
-
-
-
-
REAL
Y
Y
Y
Y
Y
Y
Y
-
-
-
-
-
-
-
-
-
-
-
DOUBLE
Y
Y
Y
Y
Y
Y
Y
-
-
-
-
-
-
-
-
-
-
-
FLOAT
Y
Y
Y
Y
Y
Y
Y
-
-
-
-
-
-
-
-
-
-
-
CHAR
Y
Y
Y
Y
-
-
-
Y
Y
Y
-
-
-
Y
-
Y
Y
Y
VARCHAR
Y
Y
Y
Y
-
-
-
Y
Y
Y
-
-
-
Y
-
Y
Y
Y
LONG VARCHAR
-
-
-
-
-
-
-
Y
Y
Y
-
-
-
Y
-
-
-
-
CHAR FOR BIT DATA
-
-
-
-
-
-
-
-
-
-
Y
Y
Y
Y
Y
-
-
-
VARCHAR FOR BIT DATA
-
-
-
-
-
-
-
-
-
-
Y
Y
Y
Y
Y
-
-
-
LONG VARCHAR FOR BIT DATA
-
-
-
-
-
-
-
-
-
-
Y
Y
Y
Y
Y
-
-
-
CLOB
-
-
-
-
-
-
-
Y
Y
Y
-
-
-
Y
-
-
-
-
BLOB
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Y
-
-
-
DATE
-
-
-
-
-
-
-
Y
Y
-
-
-
-
-
-
Y
-
Y
TIME
-
-
-
-
-
-
-
Y
Y
-
-
-
-
-
-
-
Y
Y
TIME STAMP
-
-
-
-
-
-
-
Y
Y
-
-
-
-
-
-
Y
Y
Y
If a conversion is valid, CASTs are allowed. Size incompatibilities between the source and target types might cause runtime errors.
Notes
In this discussion, the Derby SQL-92 data types are categorized as follows:
 
numeric
 
exact numeric (SMALLINT, INTEGER, BIGINT, DECIMAL, NUMERIC)
 
approximate numeric (FLOAT, REAL, DOUBLE PRECISION)
 
string
 
character string (CLOB, CHAR, VARCHAR, LONG VARCHAR )
 
bit string (BLOB, CHAR FOR BIT DATA, VARCHAR FOR BIT DATA, LONG VARCHAR FOR BIT DATA )
 
date/time
 
 
 
Conversions from numeric types
A numeric type can be converted to any other numeric type. If the target type cannot represent the non-fractional component without truncation, an exception is raised. If the target numeric cannot represent the fractional component (scale) of the source numeric, then the source is silently truncated to fit into the target. For example, casting 763.1234 as INTEGER yields 763.
Conversions from and to bit strings
Bit strings can be converted to other bit strings, but not character strings. Strings that are converted to bit strings are padded with trailing zeros to fit the size of the target bit string. The BLOB type is more limited and requires explicit casting. In most cases the BLOB type cannot be casted to and from other types.
Conversions of date/time values
A date/time value can always be converted to and from a TIMESTAMP. If a DATE is converted to a TIMESTAMP, the TIME component of the resulting TIMESTAMP is always 00:00:00. If a TIME data value is converted to a TIMESTAMP, the DATE component is set to the value of CURRENT_DATE at the time the CAST is executed. If a TIMESTAMP is converted to a DATE, the TIME component is silently truncated. If a TIMESTAMP is converted to a TIME, the DATE component is silently truncated.
SELECT CAST (miles AS INT) FROM Flights -- convert timestamps to text INSERT INTO mytable (text_column) VALUES (CAST (CURRENT_TIMESTAMP AS VARCHAR(100))) -- you must cast NULL as a data type to use it SELECT airline FROM Airlines UNION ALL VALUES (CAST (NULL AS CHAR(2))) -- cast a double as a decimal SELECT CAST (FLYING_TIME AS DECIMAL(5,2)) FROM FLIGHTS -- cast a SMALLINT to a BIGINT VALUES CAST (CAST (12 as SMALLINT) as BIGINT)
CHAR
The CHAR function returns a fixed-length character string representation of:
 
a character string, if the first argument is any type of character string.
 
a datetime value, if the first argument is a date, time, or timestamp.
 
a decimal number, if the first argument is a decimal number.
 
a double-precision floating-point number, if the first argument is a DOUBLE or REAL.
 
an integer number, if the first argument is a SMALLINT, INTEGER, or BIGINT.
The first argument must be of a built-in data type. The result of the function is a fixed-length character string. If the first argument can be null, the result can be null. If the first argument is null, the result is the null value.
Character to character syntax
CHAR (CharacterExpression [, integer] )
CharacterExpression
An expression that returns a value that is CHAR, VARCHAR, LONG VARCHAR, or CLOB data type.
integer
The length attribute for the resulting fixed length character string. The value must be between 0 and 254.
If the length of the character-expression is less than the length attribute of the result, the result is padded with blanks up to the length of the result. If the length of the character-expression is greater than the length attribute of the result, truncation is performed. A warning is returned unless the truncated characters were all blanks and the character-expression was not a long string (LONG VARCHAR or CLOB).
Integer to character syntax
CHAR (IntegerExpression )
IntegerExpression
An expression that returns a value that is an integer data type (either SMALLINT, INTEGER or BIGINT).
The result is the character string representation of the argument in the form of an SQL integer constant. The result consists of n characters that are the significant digits that represent the value of the argument with a preceding minus sign if the argument is negative. It is left justified.
 
If the first argument is a small integer: The length of the result is 6. If the number of characters in the result is less than 6, then the result is padded on the right with blanks to length 6.
 
If the first argument is a large integer: The length of the result is 11. If the number of characters in the result is less than 11, then the result is padded on the right with blanks to length 11.
 
If the first argument is a big integer: The length of the result is 20. If the number of characters in the result is less than 20, then the result is padded on the right with blanks to length 20.
Datetime to character syntax
CHAR (DatetimeExpression )
DatetimeExpression
An expression that is one of the following three data types:
 
date: The result is the character representation of the date. The length of the result is 10.
 
time: The result is the character representation of the time. The length of the result is 8.
 
timestamp: The result is the character string representation of the timestamp. The length of the result is 26.
Decimal to character
CHAR (DecimalExpression )
DecimalExpression
An expression that returns a value that is a decimal data type. If a different precision and scale is desired, the DECIMAL scalar function can be used first to make the change.
Floating point to character syntax
CHAR (FloatingPointExpression )
FloatingPointExpression
An expression that returns a value that is a floating-point data type (DOUBLE or REAL).
Use the CHAR function to return the values for EDLEVEL (defined as smallint) as a fixed length character string:
SELECT CHAR(EDLEVEL) FROM EMPLOYEE
An EDLEVEL of 18 would be returned as the CHAR(6) value '18 ' (18 followed by four blanks).
LENGTH
LENGTH is applied to either a character string expression or a bit string expression and returns the number of characters in the result.
Because all built-in data types are implicitly converted to strings, this function can act on all built-in data types.
Syntax
LENGTH ( { CharacterExpression | BitExpression } )
-- returns 20 VALUES LENGTH('supercalifragilistic') -- returns 1 VALUES LENGTH(X'FF') -- returns 4 VALUES LENGTH(1234567890)
Concatenation
The concatenation operator, ||, concatenates its right operand to the end of its left operand. It operates on a character or bit expression.
Because all built-in data types are implicitly converted to strings, this function can act on all built-in data types.
Syntax
{ { CharacterExpression || CharacterExpression } | { BitExpression || BitExpression } }
For character strings, if both the left and right operands are of type CHAR, the resulting type is CHAR; otherwise, it is VARCHAR. The normal blank padding/trimming rules for CHAR and VARCHAR apply to the result of this operator.
The length of the resulting string is the sum of the lengths of both operands.
For bit strings, if both the left and the right operands are of type CHAR FOR BIT DATA, the resulting type is CHAR FOR BIT DATA; otherwise, it is VARCHAR FOR BIT DATA.
--returns 'supercalifragilisticexbealidocious(sp?)' VALUES 'supercalifragilistic' || 'exbealidocious' || '(sp?)' -- returns NULL VALUES CAST (null AS VARCHAR(7))|| 'AString' -- returns '130asdf' VALUES '130' || 'asdf'
NULLIF and CASE expressions
Use the CASE and NULLIF expressions for conditional expressions in Derby.
NULLIF expression syntax
NULLIF(L, R)
The NULLIF expression is very similar to the CASE expression. For example:
NULLIF(V1,V2)
is equivalent to the following CASE expression:
CASE WHEN V1=V2 THEN NULL ELSE V1 END
CASE expression syntax
You can place a CASE expression anywhere an expression is allowed. It chooses an expression to evaluate based on a boolean test.
CASE WHEN BooleanExpression THEN thenExpression ELSE elseExpression END
ThenExpression and ElseExpression are both expressions that must be type-compatible. For built-in types, this means that the types must be the same or a built-in broadening conversion must exist between the types.
You do not need to use the CASE expression for avoiding NullPointerExceptions when a nullable column becomes a method receiver.
-- returns 3 VALUES CASE WHEN 1=1 THEN 3 ELSE 4 END;
If the value of the instance specified in an instance method invocation is null, the result of the invocation is null (SQL NULL). However, you still might need to use the CASE expression for when a nullable column is a primitive method parameter.
COUNT
COUNT is an aggregate function that counts the number of rows accessed in an expression (see Aggregates (set functions) ). COUNT is allowed on all types of expressions.
Syntax
COUNT ( [ DISTINCT | ALL ] Expression )
The DISTINCT qualifier eliminates duplicates. The ALL qualifier retains duplicates. ALL is assumed if neither ALL nor DISTINCT is specified. For example, if a column contains the values 1, 1, 1, 1, and 2, COUNT(col) returns a greater value than COUNT(DISTINCT col).
Only one DISTINCT aggregate expression per SelectExpression is allowed. For example, the following query is not allowed:
-- query not allowed SELECT COUNT (DISTINCT flying_time), SUM (DISTINCT miles) FROM Flights
An Expression can contain multiple column references or expressions, but it cannot contain another aggregate or subquery. If an Expression evaluates to NULL, the aggregate is not processed for that value.
The resulting data type of COUNT is BIGINT .
-- Count the number of countries in each region, -- show only regions that have at least 2 SELECT COUNT (country), region FROM Countries GROUP BY region HAVING COUNT (country) > 1
COUNT(*)
COUNT(*) is an aggregate function that counts the number of rows accessed. No NULLs or duplicates are eliminated. COUNT(*) does not operate on an expression.
Syntax
COUNT(*)
The resulting data type is BIGINT .
-- Count the number of rows in the Flights table SELECT COUNT(*) FROM Flights
CURRENT DATE
CURRENT DATE is a synonym for CURRENT_DATE .
CURRENT_DATE
CURRENT_DATE returns the current date; the value returned does not change if it is executed more than once in a single statement. This means the value is fixed even if there is a long delay between fetching rows in a cursor.
Syntax
CURRENT_DATE
or, alternately
CURRENT DATE
-- find available future flights: SELECT * FROM Flightavailability where flight_date > CURRENT_DATE;
CURRENT ISOLATION
CURRENT ISOLATION returns the current isolation level as a char(2) value of either ""(blank), "UR", "CS", "RS", or "RR".
Syntax
CURRENT ISOLATION
VALUES CURRENT ISOLATION
CURRENT SCHEMA
CURRENT SCHEMA returns the schema name used to qualify unqualified database object references.
Note: CURRENT SCHEMA and CURRENT SQLID are synonyms.
These functions return a string of up to 128 characters.
Syntax
CURRENT SCHEMA -- or, alternately: CURRENT SQLID
-- Set the name column default to the current schema: CREATE TABLE mytable (id int, name VARCHAR(128) DEFAULT CURRENT SQLID) -- Inserts default value of current schema value into the table: INSERT INTO mytable(id) VALUES (1) -- Returns the rows with the same name as the current schema: SELECT name FROM mytable WHERE name = CURRENT SCHEMA
CURRENT TIME
CURRENT TIME is a synonym for CURRENT_TIME .
CURRENT_TIME
CURRENT_TIME returns the current time; the value returned does not change if it is executed more than once in a single statement. This means the value is fixed even if there is a long delay between fetching rows in a cursor.
Syntax
CURRENT_TIME
or, alternately
CURRENT TIME
VALUES CURRENT_TIME -- or, alternately: VALUES CURRENT TIME
CURRENT TIMESTAMP
CURRENT TIMESTAMP is a synonym for CURRENT_TIMESTAMP .
CURRENT_TIMESTAMP
CURRENT_TIMESTAMP returns the current timestamp; the value returned does not change if it is executed more than once in a single statement. This means the value is fixed even if there is a long delay between fetching rows in a cursor.
Syntax
CURRENT_TIMESTAMP
or, alternately
CURRENT TIMESTAMP
VALUES CURRENT_TIMESTAMP -- or, alternately: VALUES CURRENT TIMESTAMP
CURRENT_USER
CURRENT_USER returns the authorization identifier of the current user (the name of the user passed in when the user connected to the database). If there is no current user, it returns APP.
USER and SESSION_USER are synonyms.
These functions return a string of up to 128 characters.
Syntax
CURRENT_USER
VALUES CURRENT_USER
DATE
The DATE function returns a date from a value. The argument must be a date, timestamp, a positive number less than or equal to 3,652,059, a valid string representation of a date or timestamp, or a string of length 7 that is not a CLOB or LONG VARCHAR. If the argument is a string of length 7, it must represent a valid date in the form yyyynnn, where yyyy are digits denoting a year, and nnn are digits between 001 and 366, denoting a day of that year. The result of the function is a date. If the argument can be null, the result can be null; if the argument is null, the result is the null value.
The other rules depend on the data type of the argument specified:
 
If the argument is a date, timestamp, or valid string representation of a date or timestamp: The result is the date part of the value.
 
If the argument is a number: The result is the date that is n-1 days after January 1, 0001, where n is the integral part of the number.
 
If the argument is a string with a length of 7: The result is the date represented by the string.
Syntax
DATE ( expression )
This example results in an internal representation of '1988-12-25'.
VALUES DATE('1988-12-25')
DAY
The DAY function returns the day part of a value. The argument must be a date, timestamp, or a valid character string representation of a date or timestamp that is neither a CLOB nor a LONG VARCHAR. The result of the function is a large integer. If the argument can be null, the result can be null; if the argument is null, the result is the null value.
The other rules depend on the data type of the argument specified:
 
If the argument is a date, timestamp, or valid string representation of a date or timestamp: The result is the day part of the value, which is an integer between 1 and 31.
 
If the argument is a time duration or timestamp duration: The result is the day part of the value, which is an integer between -99 and 99. A nonzero result has the same sign as the argument.
Syntax
DAY ( expression )
values day('2005-08-02');
The resulting value is 2.
DOUBLE
The DOUBLE function returns a floating-point number corresponding to a:
 
number if the argument is a numeric expression.
 
character string representation of a number if the argument is a string expression.
Numeric to double
DOUBLE [PRECISION] (NumericExpression )
NumericExpression
The argument is an expression that returns a value of any built-in numeric data type.
The result of the function is a double-precision floating-point number. If the argument can be null, the result can be null; if the argument is null, the result is the null value. The result is the same number that would occur if the argument were assigned to a double-precision floating-point column or variable.
Character string to double
DOUBLE (StringExpression )
StringExpression
The argument can be of type CHAR or VARCHAR in the form of a numeric constant. Leading and trailing blanks in argument are ignored.
The result of the function is a double-precision floating-point number. The result can be null; if the argument is null, the result is the null value. The result is the same number that would occur if the string was considered a constant and assigned to a double-precision floating-point column or variable.
HOUR
The HOUR function returns the hour part of a value. The argument must be a time, timestamp, or a valid character string representation of a time or timestamp that is neither a CLOB nor a LONG VARCHAR. The result of the function is a large integer. If the argument can be null, the result can be null; if the argument is null, the result is the null value.
The other rules depend on the data type of the argument specified:
 
If the argument is a date, timestamp, or valid string representation of a date or timestamp: The result is the hour part of the value, which is an integer between 0 and 24.
 
If the argument is a time duration or timestamp duration: The result is the hour part of the value, which is an integer between -99 and 99. A nonzero result has the same sign as the argument.
Syntax
HOUR ( expression )
Select all the classes that start in the afternoon from a table called TABLE1.
SELECT * FROM TABLE1 WHERE HOUR(STARTING) BETWEEN 12 AND 17
IDENTITY_VAL_LOCAL
Derby supports the IDENTITY_VAL_LOCAL function.
Syntax:
IDENTITY_VAL_LOCAL ( )
The IDENTITY_VAL_LOCAL function is a non-deterministic function that returns the most recently assigned value of an identity column for a connection, where the assignment occurred as a result of a single row INSERT statement using a VALUES clause.
The IDENTITY_VAL_LOCAL function has no input parameters. The result is a DECIMAL (31,0), regardless of the actual data type of the corresponding identity column.
The value returned by the IDENTITY_VAL_LOCAL function, for a connection, is the value assigned to the identity column of the table identified in the most recent single row INSERT statement. The INSERT statement must contain a VALUES clause on a table containing an identity column. The assigned value is an identity value generated by Derby. The function returns a null value when a single row INSERT statement with a VALUES clause has not been issued for a table containing an identity column.
The result of the function is not affected by the following:
 
A single row INSERT statement with a VALUES clause for a table without an identity column
 
A multiple row INSERT statement with a VALUES clause
 
An INSERT statement with a fullselect
If a table with an identity column has an INSERT trigger defined that inserts into another table with another identity column, then the IDENTITY_VAL_LOCAL() function will return the generated value for the statement table, and not for the table modified by the trigger.
Examples:
ij> create table t1(c1 int generated always as identity, c2 int); 0 rows inserted/updated/deleted ij> insert into t1(c2) values (8); 1 row inserted/updated/deleted ij> values IDENTITY_VAL_LOCAL(); 1 ------------------------------- 1 1 row selected ij> select IDENTITY_VAL_LOCAL()+1, IDENTITY_VAL_LOCAL()-1 from t1; 1 |2 ------------------------------------------------------------------- 2 |0 1 row selected ij> insert into t1(c2) values (IDENTITY_VAL_LOCAL()); 1 row inserted/updated/deleted ij> select * from t1; C1 |C2 ------------------------------- 1 |8 2 |1 2 rows selected ij> values IDENTITY_VAL_LOCAL(); 1 ------------------------------- 2 1 row selected ij> insert into t1(c2) values (8), (9); 2 rows inserted/updated/deleted ij> -- multi-values insert, return value of the function should not change values IDENTITY_VAL_LOCAL(); 1 ------------------------------- 2 1 row selected ij> select * from t1; C1 |C2 ------------------------------- 1 |8 2 |1 3 |8 4 |9 4 rows selected ij> insert into t1(c2) select c1 from t1; 4 rows inserted/updated/deleted -- insert with sub-select, return value should not change ij> values IDENTITY_VAL_LOCAL(); 1 ------------------------------- 2 1 row selected ij> select * from t1; C1 |C2 ------------------------------- 1 |8 2 |1 3 |8 4 |9 5 |1 6 |2 7 |3 8 |4 8 rows selected
INTEGER
The INTEGER function returns an integer representation of a number, character string, date, or time in the form of an integer constant.
Syntax
INT[EGER] (NumericExpression | CharacterExpression )
NumericExpression
An expression that returns a value of any built-in numeric data type. If the argument is a numeric-expression, the result is the same number that would occur if the argument were assigned to a large integer column or variable. If the whole part of the argument is not within the range of integers, an error occurs. The decimal part of the argument is truncated if present.
CharacterExpression
An expression that returns a character string value of length not greater than the maximum length of a character constant. Leading and trailing blanks are eliminated and the resulting string must conform to the rules for forming an SQL integer constant. The character string cannot be a long string. If the argument is a character-expression, the result is the same number that would occur if the corresponding integer constant were assigned to a large integer column or variable.
The result of the function is a large integer. If the argument can be null, the result can be null; if the argument is null, the result is the null value.
Using the EMPLOYEE table, select a list containing salary (SALARY) divided by education level (EDLEVEL). Truncate any decimal in the calculation. The list should also contain the values used in the calculation and employee number (EMPNO). The list should be in descending order of the calculated value:
SELECT INTEGER (SALARY / EDLEVEL), SALARY, EDLEVEL, EMPNO FROM EMPLOYEE ORDER BY 1 DESC
LOCATE
If a specified substring is found within a specified search string, LOCATE returns the index at which the substring is found within the search string. If the substring is not found, LOCATE returns 0.
Syntax
LOCATE(CharacterExpression, CharacterExpression [, StartPosition] )
The second CharacterExpression is the search string and is searched from the beginning, unless startPosition is specified, in which case the search begins from position there; the index starts with 1. It returns 0 if the string is not found.
The return type for LOCATE is an integer.
-- returns 2 VALUES LOCATE('love', 'clover')
LCASE or LOWER
LCASE or LOWER takes a character expression as a parameter and returns a string in which all alpha characters have been converted to lowercase.
Syntax
LCASE or LOWER ( CharacterExpression )
A CharacterExpression is a CHAR, VARCHAR, or LONG VARCHAR data type or any built-in type that is implicitly converted to a string (except a bit expression).
If the parameter type is CHAR or LONG VARCHAR, the return type is CHAR or LONG VARCHAR. Otherwise, the return type is VARCHAR.
The length and maximum length of the returned value are the same as the length and maximum length of the parameter.
If the CharacterExpression evaluates to null, this function returns null.
-- returns 'asd1#w' VALUES LOWER('aSD1#w') SELECT LOWER(flight_id) FROM Flights
LTRIM
LTRIM removes blanks from the beginning of a character string expression.
Syntax
LTRIM(CharacterExpression)
A CharacterExpression is a CHAR, VARCHAR, or LONG VARCHAR data type, any built-in type that is implicitly converted to a string.
LTRIM returns NULL if CharacterExpression evaluates to null.
-- returns 'asdf ' VALUES LTRIM(' asdf ')
MAX
MAX is an aggregate function that evaluates the maximum of the expression over a set of values (see Aggregates (set functions) ). MAX is allowed only on expressions that evaluate to built-in data types (including CHAR, VARCHAR, DATE, TIME, CHAR FOR BIT DATA, etc.).
Syntax
MAX ( [ DISTINCT | ALL ] Expression )
The DISTINCT qualifier eliminates duplicates. The ALL qualifier retains duplicates. These qualifiers have no effect in a MAX expression. Only one DISTINCT aggregate expression per SelectExpression is allowed. For example, the following query is not allowed:
SELECT COUNT (DISTINCT flying_time), MAX (DISTINCT miles) FROM Flights
The Expression can contain multiple column references or expressions, but it cannot contain another aggregate or subquery. It must evaluate to a built-in data type. You can therefore call methods that evaluate to built-in data types. (For example, a method that returns a java.lang.Integer or int evaluates to an INTEGER.) If an expression evaluates to NULL, the aggregate skips that value.
For CHAR, VARCHAR, and LONG VARCHAR , the number of blank spaces at the end of the value can affect how MAX is evaluated. For example, if the values 'z' and 'z ' are both stored in a column, you cannot control which one will be returned as the maximum, because a blank space has no value.
The resulting data type is the same as the expression on which it operates (it will never overflow).
-- find the latest date in the FlightAvailability table SELECT MAX (flight_date) FROM FlightAvailability -- find the longest flight originating from each airport, -- but only when the longest flight is over 10 hours SELECT MAX(flying_time), orig_airport FROM Flights GROUP BY orig_airport HAVING MAX(flying_time) > 10
MIN
MIN is an aggregate expression that evaluates the minimum of an expression over a set of rows (see Aggregates (set functions) ). MIN is allowed only on expressions that