SQL Server - Bài 4

SQL Server

T-SQL for Data Definition

Vu Tuyet Trinh

trinhvt-fit@mail.hut.edu.vn

Hanoi University of Technology

1

Overview of Transact-SQL

 Based on AINSI SQL 92 standard

 Composing of three categories

 Data Manipulation Language (DML)

 Data Definition Language (DDL)

 Data Control Language (DCL)

 Having some Microsoft specific extensions

 Beyond relational data

 net framework integration

Overview of Database Objects

Creating a New Database

 Factors to consider

 Default: Sysadmin, dbcreator

 Creator becomes the owner

 Maximum of 32,767 per server

 Follow naming rules

Creating a New Database

 Some arguments:

 The name of the database

 The size of the database

 The files where the database will reside

CREATE DATABASE Sample ON

PRIMARY ( NAME=SampleData, FILENAME='c:\Program Files\ \ \Data\Sample.mdf', SIZE=10MB,

MAXSIZE=15MB, FILEGROWTH=20%) LOG ON

( NAME=SampleLog, FILENAME= 'c:\Program Files\ \ \Data\Sample.ldf', SIZE=3MB,

MAXSIZE=5MB, FILEGROWTH=1MB) COLLATE SQL_Latin1_General_Cp1_CI_AS

CREATE DATABASE Sample ON

PRIMARY ( NAME=SampleData, FILENAME='c:\Program Files\ \ \Data\Sample.mdf', SIZE=10MB,

MAXSIZE=15MB, FILEGROWTH=20%) LOG ON

( NAME=SampleLog, FILENAME= 'c:\Program Files\ \ \Data\Sample.ldf', SIZE=3MB,

MAXSIZE=5MB, FILEGROWTH=1MB) COLLATE SQL_Latin1_General_Cp1_CI_AS

Setting & Viewing Database Options

 Set Database Options By Using:

 SQL Server Management Studio

 ALTER DATABASE statement

 Database Option Categories

Retrieving Database Information

 Determining database properties by using the

(‘pubs’,’useraccess’)

 Using system stored procedures to display

information about databases and its parameters

Attaching an Existing Database

Creating a Snapshot Database

Managing Databases

 Managing Data and Log File Growth

 Monitoring and Expanding a Transaction Log

 Shrinking a Database or File

 Dropping a Database

Managing Data and Log File Growth

ALTER DATABASE Sample MODIFY FILE ( NAME = 'SampleLog', SIZE = 15MB)

GO

ALTER DATABASE Sample ADD FILE

(NAME = SampleData2, FILENAME='c:\Program Files\ \ \

Data\Sample2.ndf', SIZE=15MB,

MAXSIZE=20MB) GO

ALTER DATABASE Sample MODIFY FILE ( NAME = 'SampleLog', SIZE = 15MB)

GO

ALTER DATABASE Sample ADD FILE

(NAME = SampleData2, FILENAME='c:\Program Files\ \ \

Data\Sample2.ndf', SIZE=15MB,

MAXSIZE=20MB) GO

 Using Automatic File Growth

 Expanding Database Files

 Adding Secondary Database Files

Monitoring and Expanding a Transaction Log

 Monitoring the log

 Monitoring situations that produce extensive log activity

 Mass loading of data into indexed table

 Large transactions

 Performing logged text or image operations

 Expanding the log when necessary

Shrinking a Database or File

 Shrinking an Entire Database

 Shrinking a Data File in the Database

 Shrinking a Database Automatically

Set autoshrink database option to true

DBCC SHRINKDATABASE (Sample, 25)

DBCC SHRINKFILE (Sample_Data, 10)

Dropping a Database

DROP DATABASE Northwind, pubs

 Methods of Dropping a Database

 SQL Server Enterprise Manager

 DROP DATABASE statement

 Restrictions on Dropping a Database

 While it is being restored

 When a user is connected to it

 When publishing as part of replication

 If it is a system database

 char, varchar, text

 Unicode character string

 nchar, nvarchar, ntext

 Binary character string

 binary, varbinary, image

 Others

 sql_variant, timestamp, xml…

User-defined Data Type

 Extending the SQL type system as

 alias data types that consist of a single SQL Server system data type or

 structure of multiple data types having behaviors (CLR Type)

 Allowing relationships among data elements within a table

 Representing a position in a hierarchy

 Storing values that represent nodes in a hierarchy tree

 Implemented as a CLR UDT

CREATE TABLE employee

( EmployeeID int NOT NULL,

EmpName varchar(20) NOT NULL,

Title varchar(20) NULL,

Salary decimal(18, 2) NOT NULL,

hireDate datetimeoffset(0) NOT NULL,

OrgNode HIERARCHYID NOT NULL )

 Methods for creating and operating on hierarchy nodes

GetAncestorGetRoot

IsDescendantParse

GetDescendantGetLevel

ReparentTostring

Spatial Data Types

 Storing geographic locations and shapes such as landmarks, roads, building, …

 Based on vector model specified in Known Text (WKT) or Known Binary (WKB) format (recommended by Open Geospatial Consortium)

Geography & Geometry Data Type

Performing Spatial Operations

 Both types provide static and instance methods

 Calculate distances, find intersections, etc

XML Data Type

 Used to store XML documents or fragments of XML

 Possibility to register XML schemas and store schema

information within the database

 Automatic validation of XML documents

 Automatic shredding of the XML data to support efficient querying and updating of the content

 Searching and updating via an implementation of XQuery and DML

 Declaration is (mostly) like other data types

 Table: CREATE TABLE T(MyXml XML)

 Variable: DECLARE @MyXml XML

 Usage is (mostly) like other data types

 Query: SELECT MyXml from T

 Variable: SET @MyXml = ‘<size>12</size>’

 Some of the limitations:

 Does not support converting to text or ntext.

 Cannot be compared or sorted So, no GROUP BY, PRIMARY KEY, etc.

 Cannot be used as a parameter to any scalar, built-in functions other than ISNULL, COALESCE, and DATALENGTH.

 Stored semantically The order of attributes and insignificant white

spaces are not preserved

Creating XML Data

 Converting Strings

 Most common way to create XML

 E.g SET @MyXml = ‘<size>12</size>’

 Bulk Loading using OPENROWSET

 Easiest way to load XML from a file

 FOR XML clause in SELECT

 Easiest way to convert from table data

XML Schema Collections

 A schema is a description of a type of XML document (e.g order)

 Schema collections validate XML vs schema (XSD)

 Creating

 CREATE XML SCHEMA COLLECTION <xsd>

 Can use either a string or xml variable for xsd

 Managing can be tricky

 ALTER ADD for new schemas, elements

 DROP to delete them Can’t have any dependencies.

 Some of the limitations

 Lax validation, xsd:include, xsd:key, xsd:keyref not supported.

 Time zone always normalized to GMT.

 Very large schemas will cause errors due to stack limits

XML Data Type Methods

 XML Data Type is the only type that supports operations

 Query() – returns XML from XQuery expression

 Value() – returns SQL type from XQuery

 Exist() – bool if XQuery exists

 Modify() – changes XML using XQuery

 Nodes() – shreds XML into relational data; think of it as a table valued split

Storing Document & Multimedia

• Low cost per GB

• Streaming Performance

• Complex application development &

deployment

• Integration with structured data

• File size limitations

• Highest cost per GB

• Lower cost per GB at scale

• Scalability &

Expandability

• Complex application development &

deployment

• Separate data management

• Enterprise-scales only

Example • Windows File Servers

• NetApp NetFiler • EMC Centera

• Fujitsu Nearline • SQL Server

VARBINARY(MAX)

Use File Servers

Use File Servers

DB

Application

BLOBs

Dedicated BLOB Store

Dedicated BLOB Store

DB

Application

BLOBs

Store BLOBs in Database

Store BLOBs in Database

DB

Application

BLOBs

BLOB Storage in SQL Server

Remote BLOB Storage SQL BLOB FILESTREAM Storage

Use File Servers

Use File Servers

DB

Application

BLOB

Dedicated BLOB Store

Dedicated BLOB Store

D B

D B

Store BLOBs

in DB + File System

Application

BLOB

DB

FILESTREAM Data Type

 Storage Attribute on VARBINARY(MAX)

 Unstructured data stored directly in the file

 Size limit is the file system volume size

 SQL Server Security Stack

Store BLOBs in DB + File System

Store BLOBs in DB + File System

Application

BLOB

DB

BLOB Storage Vision

File Stores / BLOB Stores

SQL BLOBs Remote BLOB

Store API FILESTREAM

Streaming

Performance Depends on

external store external store Depends on

Link Level Consistency

Data Level

Consistency

Integrated Management

Non-local Windows

Interop with External

Creating Tables

 Determining column & data type

 Determining column nullability

 Defining column default values

CREATE TABLE dbo.Categories

nvarchar(15) ntext

image

int IDENTITY (1,1)

nvarchar(15) ntext

image

NOT NULL, NOT NULL, NULL,

NULL)

NOT NULL, NOT NULL, NULL,

NULL)

Modifying Table Definition

ALTER TABLE table_name

{[ALTER COLUMN column_name

|DROP

{[ CONSTRAINT ] constraint_name | COLUMN column }

]}

ALTER TABLE CategoriesNew

DROP COLUMN Sales_date

ALTER TABLE CategoriesNew

DROP COLUMN Sales_date

DROP

Customer_name Sales_amount Sales_date Customer ID Commission

Views

Summary

 Data definition language

 Creating and managing database object

Create, Use, Alter, Drop,

 SQL Server database objects

database, table, view, index, …

 Simplifying the management of more complicated data

Microsoft