Tutorial sobre SQL (DDL, DML) no exemplo do dialeto MS SQL Server

Introdução

• SELECIONAR – seleção de dados
• INSERIR – inserção de novos dados
• ATUALIZAÇÃO – atualização de dados
• EXCLUIR – exclusão de dados
• MERGE – mesclagem de dados

Resumidamente sobre MS SQL Server Management Studio (SSMS)

Teoria

• A linguagem DDL serve para criar e modificar uma estrutura de banco de dados (exclusão de tabelas e links);
• A linguagem DML permite manipular os dados da tabela, suas linhas. Ele também serve para selecionar dados de tabelas, adicionar novos dados, bem como atualizar e excluir dados atuais.

-- comentário de linha única

/* comentário delimitado */

DDL – Linguagem de definição de dados

• ID do funcionário – inteiro
• Nome Completo – string
• Data de nascimento – data
• E-mail – string
• Posição – string
• Departamento – string

CRIAR BANCO DE DADOS

Teste DROP DATABASE

USE Teste

CREATE TABLE [Employees]( [EmployeeID] int, [FullName] nvarchar(30), [Birthdate] date, [E-mail] nvarchar(30), [Position] nvarchar(30), [Department] nvarchar( 30) )

DROP TABLE [Funcionários]

• ID
• Nome
• Aniversário
• E-mail
• Posição
• Departamento

CREATE TABLE Employees( ID int, Name nvarchar(30), Data de aniversário, Email nvarchar(30), Cargo nvarchar(30), Department nvarchar(30) ) 

-- Campo de ID updateALTER TABLE Funcionários ALTER COLUMN ID int NOT NULL-- Campo de nome updateALTER TABLE Funcionários ALTER COLUMN Nome nvarchar(30) NOT NULL

-- create table CREATE TABLE Employees( ID int, -- No ORACLE o tipo int é um valor para number(38) Name nvarchar2(30), -- no ORACLE nvarchar2 é idêntico ao nvarchar no MS SQL Data de aniversário, Email nvarchar2(30), Posição nvarchar2(30), Departamento nvarchar2(30) ); -- Atualização do campo ID e Nome (aqui usamos MODIFY(…) ao invés de ALTER COLUMNALTER TABLE Funcionários MODIFY(ID int NOT NULL,Name nvarchar2(30) NOT NULL); -- adiciona PK (neste caso a construção é a mesma como no MS SQL) ALTER TABLE Funcionários ADD CONSTRAINT PK_Employees PRIMARY KEY(ID);

NAME varchar2(30 BYTE) – capacidade do campo é igual a 30 bytes NAME varchar2(30 CHAR) – capacidade do campo é igual a 30 símbolos

INSERIR Funcionários(ID,Cargo,Departamento) VALORES (1000,'CEO,N'Administração'), (1001,N'Programador',N'IT'), (1002,N'Contador',N'Contas dept'), (1003,N'Programador Sênior',N'IT')

INSERIR Funcionários(ID,Cargo,Departamento,Nome) VALORES (1000,N'CEO',N'Administração',N'John'), (1001,N'Programador',N'IT',N'Daniel '), (1002,N'Contador',N'Departamento de Contas',N'Mike'), (1003,N'Programador Sênior',N'IT',N'Jordan')

Funcionários DROP TABLE

CREATE TABLE Employees( ID int NOT NULL, Name nvarchar(30) NOT NULL, Data de aniversário, Email nvarchar(30), Cargo nvarchar(30), Department nvarchar(30) )

ALTER TABLE Funcionários ALTER COLUMN Nome nvarchar(30) NULL

ALTER TABLE Funcionários ALTER COLUMN Nome nvarchar(30)

ALTER TABLE Funcionários ALTER COLUMN Nome nvarchar(50)

Chave primária

ALTER TABLE Funcionários ADD CONSTRAINT PK_Employees PRIMARY KEY(ID)

ALTER TABLE table_name ADD CONSTRAINT constraint_name PRIMARY KEY(field1,field2,…)

Funcionários DROP TABLE

CREATE TABLE Employees( ID int NOT NULL, Name nvarchar(30) NOT NULL, Data de aniversário, Email nvarchar(30), Posição nvarchar(30), Department nvarchar(30), CONSTRAINT PK_Employees PRIMARY KEY(ID) – descrever PK após todos os campos como restrição)

INSERIR Funcionários(ID,Cargo,Departamento,Nome) VALORES (1000,N'CEO',N'Administração',N'John'), (1001,N'Programador',N'IT',N'Daniel '), (1002,N'Contador',N'Departamento de contas',N'Mike'), (1003,N'Programador sênior',N'IT',N'Jordan')

CREATE TABLE Employees( ID int NOT NULL, Name nvarchar(30) NOT NULL, Data de aniversário, Email nvarchar(30), Posição nvarchar(30), Department nvarchar(30), PRIMARY KEY(ID) )

CREATE TABLE Employees( ID int NOT NULL PRIMARY KEY, Name nvarchar(30) NOT NULL, Data de aniversário, Email nvarchar(30), Posição nvarchar(30), Department nvarchar(30) )

ALTER TABLE Funcionários DROP CONSTRAINT PK_Employees

Resumo:

• CRIAR TABELA table_name (lista de campos e seus tipos, bem como restrições) – serve para criar uma nova tabela no banco de dados atual;
• DERRAR TABELA table_name – serve para deletar uma tabela do banco de dados atual;
• ALTER TABLE table_name ALTER COLUMN column_name … – serve para atualizar o tipo de coluna ou para modificar suas configurações (por exemplo, quando você precisa definir NULL ou NOT NULL);
• ALTER TABLE table_name ADICIONAR CONSTRAINT constraint_name CHAVE PRIMÁRIA (campo1, campo2,…) – usado para adicionar uma chave primária à tabela atual;
• ALTER TABLE table_name DROP CONSTRAINT constraint_name – usado para excluir uma restrição da tabela.

Tabelas temporárias

CREATE TABLE #Temp( ID int, Nome nvarchar(30) )

DROP TABLE #Temp

SELECT ID,Name INTO #Temp FROM Employees

Normalização de banco de dados:divisão em subtabelas (tabelas de referência) e definição de relacionamentos de tabela

CREATE TABLE Positions( ID int IDENTITY(1,1) NOT NULL CONSTRAINT PK_Positions PRIMARY KEY, Name nvarchar(30) NOT NULL ) CREATE TABLE Departamentos( ID int IDENTITY(1,1) NOT NULL CONSTRAINT PK_Departments PRIMARY KEY, Name nvarchar(30) NOT NULL )

-- preencha o campo Nome da tabela Cargos com valores exclusivos do campo Cargo da tabela Funcionários INSERT Cargos(Nome) SELECT DISTINCT Cargo FROM Funcionários WHERE Cargo NÃO É NULO – descarta registros onde um cargo não é especificado 
 Você precisa fazer as mesmas etapas para a tabela Departamentos:
 
INSERIR Departamentos(Nome) SELECT Departamento DISTINTO DE Funcionários ONDE Departamento NÃO É NULO
 
 Agora, se abrirmos as tabelas Cargos e Departamentos, veremos uma lista numerada de valores no campo ID:
 
SELECIONAR * DE Posições
 
ID 
 
Nome 
 
 
1
 
Contador
 
 
2
 
CEO
 
 
3
 
Programador
 
 
4
 
Programador Sênior
 
 
 
 
SELECIONAR * DE Departamentos
 
ID 
 
Nome 
 
 
1
 
Administração
 
 
2
 
Departamento de contas
 
 
3
 
TI
 
 
 
 
 Essas tabelas serão as tabelas de referência para definir cargos e departamentos. Agora, vamos nos referir a identificadores de cargos e departamentos. Primeiramente, vamos criar novos campos na tabela Employees para armazenar os identificadores:
 
-- adiciona um campo para o ID cargo ALTER TABLE Employees ADD PositionID int -- adiciona um campo para o ID departamento ALTER TABLE Employees ADD DepartmentID int
 
 O tipo de campos de referência deve ser o mesmo das tabelas de referência, neste caso, é int. 
 
 Além disso, você pode adicionar vários campos usando um comando listando os campos separados por vírgulas:
 
ALTER TABLE Funcionários ADD PositionID int, DepartmentID int
 
 Agora, adicionaremos restrições de referência (FOREIGN KEY) a esses campos, para que um usuário não possa adicionar valores que não sejam os valores de ID das tabelas de referência. 
 
ALTER TABLE Funcionários ADICIONAR CONSTRAINT FK_Employees_PositionID FOREIGN KEY(PositionID) REFERENCES Positions(ID)
 
 Os mesmos passos devem ser feitos para o segundo campo:
 
ALTER TABLE Funcionários ADICIONAR CONSTRAINT FK_Employees_DepartmentID FOREIGN KEY(DepartmentID) REFERENCES Departments(ID)
 
 Agora, os usuários só podem inserir nesses campos os valores de ID da tabela de referência correspondente. Assim, para utilizar um novo departamento ou cargo, o usuário deve adicionar um novo registro na tabela de referência correspondente. Como os cargos e departamentos são armazenados em tabelas de referência em uma cópia, para alterar seu nome, você precisa alterá-lo apenas na tabela de referência. 
 
 O nome de uma restrição de referência geralmente é composto. Consiste no prefixo «FK» seguido de um nome de tabela e um nome de campo que se refere ao identificador da tabela de referência. 
 
 O identificador (ID) geralmente é um valor interno usado apenas para links. Não importa o valor que tenha. Assim, não tente se livrar de lacunas na sequência de valores que aparecem quando você trabalha com a tabela, por exemplo, quando você exclui registros da tabela de referência. 
 
 Em alguns casos, é possível construir uma referência a partir de vários campos:
 
ALTER TABLE tabela ADD CONSTRAINT constraint_name FOREIGN KEY(campo1,campo2,…) REFERENCES tabela de referência(campo1,campo2,…)
 
 Neste caso, uma chave primária é representada por um conjunto de vários campos (campo1, campo2, …) na tabela “reference_table”. 
 
 Agora, vamos atualizar os campos PositionID e DepartmentID com os valores de ID das tabelas de referência. 
 
 Para fazer isso, usaremos o comando UPDATE:
 
UPDATE e SET PositionID=(SELECT ID FROM Cargos WHERE Nome=e.Position), DepartmentID=(SELECT ID FROM Departments WHERE Name=e.Department) FROM Employees e
 
 Execute a seguinte consulta:
 
SELECIONAR * DE Funcionários
 
ID 
 
Nome 
 
Aniversário 
 
E-mail 
 
Posição 
 
Departamento 
 
ID da posição 
 
DepartmentID 
 
 
1000
 
João
 
NULO
 
NULO
 
CEO
 
Administração
 
2
 
1
 
 
1001
 
Daniel
 
NULO
 
NULO
 
Programador
 
TI
 
3
 
3
 
 
1002
 
Mike
 
NULO
 
NULO
 
Contador
 
Departamento de contas
 
1
 
2
 
 
1003
 
Jordânia
 
NULO
 
NULO
 
Programador sênior
 
TI
 
4
 
3
 
 
 
 
 Como você pode ver, os campos PositionID e DepartmentID correspondem a cargos e departamentos. Assim, você pode excluir os campos Cargo e Departamento da tabela Funcionários executando o seguinte comando:
 
ALTER TABLE Funcionários DROP COLUMN Cargo,Departamento
 
 Agora, execute esta declaração:
 
SELECIONAR * DE Funcionários
 
ID 
 
Nome 
 
Aniversário 
 
E-mail 
 
PositionID 
 
DepartmentID 
 
 
1000
 
João
 
NULO
 
NULO
 
2
 
1
 
 
1001
 
Daniel
 
NULO
 
NULO
 
3
 
3
 
 
1002
 
Mike
 
NULO
 
NULO
 
1
 
2
 
 
1003
 
Jordânia
 
NULO
 
NULO
 
4
 
3
 
 
 
 
 Portanto, não temos sobrecarga de informações. Podemos definir os nomes de cargos e departamentos por seus identificadores usando os valores nas tabelas de referência:
 
SELECT e.ID,e.Name,p.Name PositionName,d.Name DepartmentName FROM Employees e LEFT JOIN Departamentos d ON d.ID=e.DepartmentID LEFT JOIN Posições p ON p.ID=e.PositionID 
ID 
 
Name 
 
PositionName 
 
DepartmentName 
 
 
1000
 
John
 
CEO
 
Administration
 
 
1001
 
Daniel
 
Programmer
 
IT
 
 
1002
 
Mike
 
Accountant
 
Accounts dept
 
 
1003
 
Jordan
 
Senior programmer
 
IT
 
 
 
 
 In the object inspector, we can see all the objects created for this table. Here we can also manipulate these objects in different ways, for example, rename or delete the objects. 
 
  
 
 In addition, it should be noted that it is possible to create a recursive reference. 
 
 Let’s consider this particular example. 
 
 Let’s add the ManagerID field to the table with employees. This new field will define an employee to whom this employee is subordinated. 
 
ALTER TABLE Employees ADD ManagerID int
 
 This field permits the NULL value as well. 
 
 Now, we will create a FOREIGN KEY for the Employees table:
 
ALTER TABLE Employees ADD CONSTRAINT FK_Employees_ManagerID FOREIGN KEY (ManagerID) REFERENCES Employees(ID)
 
 Then create a diagram and check how our tables are linked:
 
  
 
 As you can see, the Employees table is linked with the Positions and Departments tables and is a recursive reference. 
 
  
 
 Finally, I would like to note that reference keys can include additional properties such as ON DELETE CASCADE and ON UPDATE CASCADE. They define the behavior when deleting or updating a record that is referenced from the reference table. If these properties are not specified, then we cannot change the ID of the record in the reference table referenced from the other table. Also, we cannot delete this record from the reference table until we remove all the rows that refer to this record or update the references to another value in these rows. 
 
 For example, let’s re-create the table and specify the ON DELETE CASCADE property for FK_Employees_DepartmentID:
 
DROP TABLE Employees CREATE TABLE Employees( ID int NOT NULL, Name nvarchar(30), Birthday date, Email nvarchar(30), PositionID int, DepartmentID int, ManagerID int, CONSTRAINT PK_Employees PRIMARY KEY (ID), CONSTRAINT FK_Employees_DepartmentID FOREIGN KEY(DepartmentID) REFERENCES Departments(ID) ON DELETE CASCADE, CONSTRAINT FK_Employees_PositionID FOREIGN KEY(PositionID) REFERENCES Positions(ID), CONSTRAINT FK_Employees_ManagerID FOREIGN KEY (ManagerID) REFERENCES Employees(ID) ) INSERT Employees (ID,Name,Birthday,PositionID,DepartmentID,ManagerID)VALUES (1000,N'John','19550219',2,1,NULL), (1001,N'Daniel','19831203',3,3,1003), (1002,N'Mike','19760607',1,2,1000), (1003,N'Jordan','19820417',4,3,1000)
 
 Let’s delete the department with identifier ‘3’ from the Departments table:
 
DELETE Departments WHERE ID=3
 
 Let’s view the data in table Employees:
 
SELECT * FROM Employees
 
ID 
 
Name 
 
Birthday 
 
Email 
 
PositionID 
 
DepartmentID 
 
ManagerID 
 
 
1000
 
John
 
1955-02-19
 
NULL
 
2
 
1
 
NULL
 
 
1002
 
Mike
 
1976-06-07
 
NULL
 
1
 
2
 
1000
 
 
 
 
 As you can see, data of Department ‘3’ has been deleted from the Employees table as well. 
 
 The ON UPDATE CASCADE property has similar behavior, but it works when updating the ID value in the reference table. For example, if we change the position ID in the Positions reference table, then DepartmentID in the Employees table will receive a new value, which we have specified in the reference table. But in this case this cannot be demonstrated, because the ID column in the Departments table has the IDENTITY property, which will not allow us to execute the following query (change the department identifier from 3 to 30):
 
UPDATE Departments SET ID=30 WHERE ID=3
 
 The main point is to understand the essence of these 2 options ON DELETE CASCADE and ON UPDATE CASCADE. I apply these options very rarely, and I recommend that you think carefully before you specify them in the reference constraint, because If an entry is accidentally deleted from the reference table, this can lead to big problems and create a chain reaction. 
 
 Let’s restore department ‘3’:
 
-- we permit to add or modify the IDENTITY value SET IDENTITY_INSERT Departments ONINSERT Departments(ID,Name) VALUES(3,N'IT') -- we prohibit to add or modify the IDENTITY value SET IDENTITY_INSERT Departments OFF
 
 We completely clear the Employees table using the TRUNCATE TABLE command:
 
TRUNCATE TABLE Employees
 
 Again, we will add data using the INSERT command:
 
INSERT Employees (ID,Name,Birthday,PositionID,DepartmentID,ManagerID)VALUES (1000,N'John','19550219',2,1,NULL), (1001,N'Daniel','19831203',3,3,1003), (1002,N'Mike','19760607',1,2,1000), (1003,N'Jordan','19820417',4,3,1000)
 
Summary:
 
 We have described the following DDL commands:
• Adding the IDENTITY property to a field allows to make this field automatically populated (count field) for the table;
• ALTER TABLE table_name ADD field_list with_features – allows you to add new fields to the table;
• ALTER TABLE table_name DROP COLUMN field_list – allows you to delete fields from the table;
• ALTER TABLE table_name ADD CONSTRAINT constraint_name FOREIGN KEY (fields) REFERENCES reference_table – allows you to determine the relationship between a table and a reference table. 
 
Other constraints – UNIQUE, DEFAULT, CHECK
 
 Using the UNIQUE constraint, you can say that the values for each row in a given field or in a set of fields must be unique. In the case of the Employees table, we can apply this restriction to the Email field. Let’s first fill the Email values, if they are not yet defined:
 
UPDATE Employees SET Email='[email protected]' WHERE ID=1000 UPDATE Employees SET Email='[email protected]' WHERE ID=1001 UPDATE Employees SET Email='[email protected]' WHERE ID=1002 UPDATE Employees SET Email='[email protected]' WHERE ID=1003
 
 Now, you can impose the UNIQUE constraint on this field:
 
ALTER TABLE Employees ADD CONSTRAINT UQ_Employees_Email UNIQUE(Email)
 
 Thus, a user will not be able to enter the same email for several employees. 
 
 The UNIQUE constraint has the following structure:the «UQ» prefix followed by the table name and a field name (after the underscore), to which the restriction applies. 
 
 When you need to add the UNIQUE constraint for the set of fields, we will list them separated by commas:
 
ALTER TABLE table_name ADD CONSTRAINT constraint_name UNIQUE(field1,field2,…)
 
 By adding a DEFAULT constraint to a field, we can specify a default value that will be inserted if, when inserting a new record, this field is not listed in the list of fields in the INSERT command. You can set this restriction when creating a table. 
 
 Let’s add the HireDate field to the Employees table and set the current date as a default value:
 
ALTER TABLE Employees ADD HireDate date NOT NULL DEFAULT SYSDATETIME()
 
 If the HireDate column already exists, then we can use the following syntax:
 
ALTER TABLE Employees ADD DEFAULT SYSDATETIME() FOR HireDate
 
 To specify the default value, execute the following command:
 
ALTER TABLE Employees ADD CONSTRAINT DF_Employees_HireDate DEFAULT SYSDATETIME() FOR HireDate
 
 As there was no such column before, then when adding it, the current date will be inserted into each entry of the HireDate field. 
 
 When creating a new record, the current date will be also automatically added, unless we explicitly specify it, i.e. specify in the list of columns. Let’s demonstrate this with an example, where we will not specify the HireDate field in the list of the values added:
 
INSERT Employees(ID,Name,Email)VALUES(1004,N'Ostin',' [email protected]')
 
 To check the result, run the command:
 
SELECT * FROM Employees
 
ID 
 
Name 
 
Birthday 
 
Email 
 
PositionID 
 
DepartmentID 
 
ManagerID 
 
HireDate 
 
 
1000
 
John
 
1955-02-19
 
[email protected]
 
2
 
1
 
NULL
 
2015-04-08
 
 
1001
 
Daniel
 
1983-12-03
 
[email protected]
 
3
 
4
 
1003
 
2015-04-08
 
 
1002
 
Mike
 
1976-06-07
 
[email protected]
 
1
 
2
 
1000
 
2015-04-08
 
 
1003
 
Jordan
 
1982-04-17
 
[email protected]
 
4
 
3
 
1000
 
2015-04-08
 
 
1004
 
Ostin
 
NULL
 
[email protected]
 
NULL
 
NULL
 
NULL
 
2015-04-08
 
 
 
 
 The CHECK constraint is used when it is necessary to check the values being inserted in the fields. For example, let’s impose this constraint on the identification number field, which is an employee ID (ID). Let’s limit the identification numbers to be in the range from 1000 to 1999:
 
ALTER TABLE Employees ADD CONSTRAINT CK_Employees_ID CHECK(ID BETWEEN 1000 AND 1999)
 
 The constraint name is usually as follows:the «CK_» prefix first followed by the table name and a field name, for which constraint is imposed. 
 
 Let’s add an invalid record to check if the constraint is working properly (we will get the corresponding error):
 
INSERT Employees(ID,Email) VALUES(2000,'[email protected]')
 
 Now, let’s change the value being inserted to 1500 and make sure that the record is inserted:
 
INSERT Employees(ID,Email) VALUES(1500,'[email protected]')
 
 We can also create UNIQUE and CHECK constraints without specifying a name:
 
ALTER TABLE Employees ADD UNIQUE(Email) ALTER TABLE Employees ADD CHECK(ID BETWEEN 1000 AND 1999)
 
 Still, this is a bad practice and it is desirable to explicitly specify the constraint name so that users can see what each object defines:
 
  
 
 A good name gives us more information about the constraint. And, accordingly, all these restrictions can be specified when creating a table, if it does not exist yet. 
 
 Let’s delete the table:
 
DROP TABLE Employees
 
 Let’s re-create the table with all the specified constraints using the CREATE TABLE command:
 
CREATE TABLE Employees( ID int NOT NULL, Name nvarchar(30), Birthday date, Email nvarchar(30), PositionID int, DepartmentID int, HireDate date NOT NULL DEFAULT SYSDATETIME(), -- I have an exception for DEFAULT CONSTRAINT PK_Employees PRIMARY KEY (ID), CONSTRAINT FK_Employees_DepartmentID FOREIGN KEY(DepartmentID) REFERENCES Departments(ID), CONSTRAINT FK_Employees_PositionID FOREIGN KEY(PositionID) REFERENCES Positions(ID), CONSTRAINT UQ_Employees_Email UNIQUE (Email), CONSTRAINT CK_Employees_ID CHECK (ID BETWEEN 1000 AND 1999) )
 
 Finally, let’s insert our employees in the table:
 
INSERT Employees (ID,Name,Birthday,Email,PositionID,DepartmentID)VALUES (1000,N'John','19550219',' [email protected] ',2,1), (1001,N'Daniel','19831203',' [email protected] ',3,3), (1002,N'Mike','19760607',' [email protected] ',1,2), (1003,N'Jordan','19820417',' [email protected]',4,3)
 
Some words about the indexes created with the PRIMARY KEY and UNIQUE constraints 
 
 When creating the PRIMARY KEY and UNIQUE constraints, the indexes with the same names (PK_Employees and UQ_Employees_Email) are automatically created. By default, the index for the primary key is defined as CLUSTERED, and for other indexes, it is set as NONCLUSTERED. 
 
 It should be noted that the clustered index is not used in all DBMSs. A table can have only one clustered (CLUSTERED) index. It means that the records of the table will be ordered by this index. In addition, we can say that this index has direct access to all the data in the table. This is the main index of the table. A clustered index can help with the optimization of queries. If we want to set the clustered index for another index, then when creating the primary key, we should specify the NONCLUSTERED property:
 
ALTER TABLE table_name ADD CONSTRAINT constraint_name PRIMARY KEY NONCLUSTERED(field1,field2,…)
 
 Let’s specify the PK_Employees constraint index as nonclustered, while the UQ_Employees_Email constraint index – as clustered. At first, delete these constraints:
 
ALTER TABLE Employees DROP CONSTRAINT PK_Employees ALTER TABLE Employees DROP CONSTRAINT UQ_Employees_Email
 
 Now, create them with the CLUSTERED and NONCLUSTERED indexes:
 
ALTER TABLE Employees ADD CONSTRAINT PK_Employees PRIMARY KEY NONCLUSTERED (ID) ALTER TABLE Employees ADD CONSTRAINT UQ_Employees_Email UNIQUE CLUSTERED (Email)
 
 Once it is done, you can see that records have been sorted by the UQ_Employees_Email clustered index:
 
SELECT * FROM Employees
 
ID 
 
Name 
 
Birthday 
 
Email 
 
PositionID 
 
DepartmentID 
 
HireDate 
 
 
1003
 
Jordan
 
1982-04-17
 
[email protected]
 
4
 
3
 
2015-04-08
 
 
1000
 
John
 
1955-02-19
 
[email protected]
 
2
 
1
 
2015-04-08
 
 
1001
 
Daniel
 
1983-12-03
 
[email protected]
 
3
 
3
 
2015-04-08
 
 
1002
 
Mike
 
1976-06-07
 
[email protected]
 
1
 
2
 
2015-04-08
 
 
 
 
 For reference tables, it is better when a clustered index is built on the primary key, as in queries we often refer to the identifier of the reference table to obtain a name (Position, Department). The clustered index has direct access to the rows of the table, and hence it follows that we can get the value of any column without additional overhead. 
 
 It is recommended that the clustered index should be applied to the fields that you use for selection very often. 
 
 Sometimes in tables, a key is created by the stubbed field. In this case, it is a good idea to specify the CLUSTERED index for an appropriate index and specify the NONCLUSTERED index when creating the stubbed field. 
 
Summary: 
 
 We have analyzed all the constraint types that are created with the «ALTER TABLE table_name ADD CONSTRAINT constraint_name …» command:
 
 
PRIMARY KEY;
 
FOREIGN KEY controls links and data referential integrity;
 
UNIQUE – serves for setting a unique value;
 
CHECK – allows monitoring the correctness of added data;
 
DEFAULT – allows specifying a default value;
 
The «ALTER TABLE  table_name DROP CONSTRAINT constraint_name» command allows deleting all the constraints.
 
 
 Additionally, we have reviewed the indexes:CLUSTERED and UNCLUSTERED. 
 
Creating unique indexes 
 
 I am going to analyze indexes created not for the PRIMARY KEY or UNIQUE constraints. 
 
 It is possible to set indexes by a field or a set of fields using the following command:
 
CREATE INDEX IDX_Employees_Name ON Employees(Name)
 
 Also, you can add the CLUSTERED, NONCLUSTERED, and UNIQUE properties as well as specify the order:ASC (by default) or DESC. 
 
CREATE UNIQUE NONCLUSTERED INDEX UQ_Employees_EmailDesc ON Employees(Email DESC)
 
 When creating the nonclustered index, the NONCLUSTERED property can be dropped as it is set by default. 
 
 To delete the index, use the command:
 
DROP INDEX IDX_Employees_Name ON Employees
 
 You can create simple indexes and constraints with the CREATE TABLE command. 
 
 At first, delete the table:
 
DROP TABLE Employees
 
 Then, create the table with all the constraints and indexes using the CREATE TABLE command:
 
CREATE TABLE Employees( ID int NOT NULL, Name nvarchar(30), Birthday date, Email nvarchar(30), PositionID int, DepartmentID int, HireDate date NOT NULL CONSTRAINT DF_Employees_HireDate DEFAULT SYSDATETIME(), ManagerID int, CONSTRAINT PK_Employees PRIMARY KEY (ID), CONSTRAINT FK_Employees_DepartmentID FOREIGN KEY(DepartmentID) REFERENCES Departments(ID), CONSTRAINT FK_Employees_PositionID FOREIGN KEY(PositionID) REFERENCES Positions(ID), CONSTRAINT FK_Employees_ManagerID FOREIGN KEY (ManagerID) REFERENCES Employees(ID), CONSTRAINT UQ_Employees_Email UNIQUE(Email), CONSTRAINT CK_Employees_ID CHECK(ID BETWEEN 1000 AND 1999), INDEX IDX_Employees_Name(Name) )
 
 Finally, add information about our employees:
 
INSERT Employees (ID,Name,Birthday,Email,PositionID,DepartmentID,ManagerID)VALUES (1000,N'John','19550219',' [email protected] ',2,1,NULL), (1001,N'Daniel','19831203',' [email protected] ',3,3,1003), (1002,N'Mike','19760607',' [email protected] ',1,2,1000), (1003,N'Jordan','19820417',' [email protected]',4,3,1000)
 
 Keep in mind that it is possible to add values with the INCLUDE command in the nonclustered index. Thus, in this case, the INCLUDE index is a clustered index where the necessary values are linked to the index, rather than to the table. These indexes can improve the SELECT query performance if there are all the required fields in the index. However, it may lead to increasing the index size, as field values are duplicated in the index. 
 
 Abstract from MSDN.  Here is how the syntax of the command to create indexes looks: 
 
CREATE [ UNIQUE ] [ CLUSTERED | NONCLUSTERED ] INDEX index_name ON