SQL Tutorial

SQL - DatabasesWhat's a Database? A SQL database is nothing more than an empty shell, like a vacant warehouse. It offers no real functionality whatsoever, but does provide a virtual space to store data. Data is stored inside of database objects called tables, and tables are the containers that actually hold specific types of data, such as numbers, files, strings, and dates.

A single database can house hundreds of tables containing more than 1,000 table columns each and they may be jam packed with relational data ready to be retrieved by SQL. Perhaps the greatest feature SQL offers is that it doesn't take much effort to rearrange your warehouse to meet your ever-growing business needs.

SQL - Creating a DatabaseCreating a database inside of SQL Express has its advantages. After launching Microsoft's SQL Server Management Studio Express application, simply right-clicking on the Databases folder of the Object Explorer gives you the option to create a New Database. After selecting the New Database... option, name your database "MyDatabase" and press "OK".

Now is the time to press the New Query button located toward the top of the screen, just above the Object Explorer pane.

Pressing this button offers an empty tab. All SQL query statements (code) that we will be exploring will be entered here and executed against the SQL Express database.

If you haven't yet created a new database, you may also create a database by typing the following SQL query statement into your new empty query tab, and then pressing the Execute button or striking the (F5) key.

SQL Create Database Query:

CREATE DATABASE MyDatabase;

After executing this query, SQL will notify you that your query has run successfully and that the database was created successfully. If you receive an error message instead, Google the error message for troubleshooting advice. (Vista users must verify that they are running SQL Server Management Studio Express with administrator privileges.)

SQL - TablesData is stored inside SQL tables which are contained within SQL databases. A single database can house hundreds of tables, each playing its own unique role in the database schema. While database architecture and schema are concepts far above the scope of this tutorial, we plan on diving in just past the surface to give you a glimpse of database architecture that begins with a thorough understanding of SQL Tables.

Advertise on Tizag.com

SQL tables are comprised of table rows and columns. Table columns are responsible for storing many different types of data, like numbers, texts, dates, and even files. There are many different types of table columns and these data types vary, depending on how the SQL table has been created by the SQL developer. A table row is a horizontal record of values that fit into each different table column.

SQL - Create a SQL TableLet's now CREATE a SQL table to help us expand our knowledge of SQL and SQL commands. This new table will serve as a practice table and we will begin to populate this table with some data which we can then manipulate as more SQL Query commands are introduced. The next couple of examples will definitely be overwhelming to novice SQL programmers, but we will take a moment to explain what's going on.

SQL Create Table Query:USE mydatabase;

CREATE TABLE orders(id INT IDENTITY(1,1) PRIMARY KEY,customer VARCHAR(50),day_of_order DATETIME,product VARCHAR(50),quantity INT);

The first line of the example, "USE mydatabase;", is pretty straightforward. This line defines the query scope and directs SQL to run the command against the MyDatabase object we created earlier in the SQL Databases lesson. The blank line break after the first command is not required, but it makes our query

http://www.tizag.com/about/advertise.php

http://sqldatabases.php/

easier to follow. The line starting with the CREATE clause is where we are actually going to tell SQL to create the new table, which is named orders.

Each table column has its own set of guidelines or schema, and the lines of code above contained in parenthesis () are telling SQL how to go about setting up each column schema. Table columns are presented in list format, and each schema is separated with a comma (,). It isn't important to fully understand exactly what all of these schema details mean just yet. They will be explained in more detail throughout the remainder of the tutorial. For now, just take note that we are creating a new, empty SQL table named orders, and this table is 5 columns wide.

SQL - INSERT DATA into your New TableNext, we will use SQL's INSERT command to draw up a query that will insert a new data row into our brand new SQL table, orders. If you're already familiar with everything we've covered so far, please execute the query below and then skip ahead and start learning about other SQL Queries.

SQL Insert Query:USE mydatabase;

INSERT INTO orders (customer,day_of_order,product, quantity)VALUES('Tizag','8/1/08','Pen',4);

SQL Insert Query Results:(1 row(s) affected)This message ("1 row(s) affected") indicates that our query has run successfully and also informs us that 1 row has been affected by the query. This is the desired result as our goal was to insert a single record into the newly formed orders table.

Listed above is a typical INSERT query used to insert data into the table we had previously created. The first line ("USE mydatabase;") identifies the query scope and the line after indicates what it is we'd like SQL to do for us. ("INSERT INTO orders") inserts data into the orders table. Then, we have to list each table column by name (customer,day_of_order,product,

http://www.tizag.com/sqlTutorial/sqlqueries.php

quantity) and finally provide a list of values to insert into each table column VALUES('Tizag','8/1/08','Pen',4).

You may notice that we have not included the id column, and this is intentional. We have set this column up in a way that allows SQL to populate this field automatically, and therefore, we do not need to worry about including it in any of our INSERT statements. (More on this later.)

SQL - QueriesSQL coins the term query as the name for its commands. Basically, all SQL code is written in the form of a query statement and then executed against a database. All SQL queries perform some type of data operation such as selecting data, inserting/updating data, or creating data objects such as SQL databases and SQL tables. Each query statement begins with a clause such as SELECT,UPDATE, CREATE or DELETE.


SELECT queries are the most commonly used SQL commands, so let's take a look at a SELECT query that will return records from the orders table that we created previously in the SQL Tables lesson.

SQL Query Code:USE mydatabase;

SELECT * FROM orders;

SQL Query Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Pen4We'll explain the mechanics of this code in the next lesson. For now, just know that SELECT queries essentially tell SQL to go and "fetch" table data for your viewing pleasure.

Here's a look at a few different query types including a INSERT and SELECT query we will be covering in the next lesson, SQL Select.

SQL Query Examples:-- Inserts data into a SQL Database/TableINSERT INTO orders (customer,day_of_order,product, quantity)VALUES('Tizag','8/1/08','Pen',4);

-- Selects data from a SQL Database/TableSELECT * FROM orders;

-- Updates data in a Database/TableUPDATE orders SET quantity = '6'

http://sqlselect.php/

http://www.tizag.com/sqlTutorial/sqltables.php


WHERE id = '1'

More information about these queries can be found at the following links: SQL Insert, SQL Update, SQL Delete

SQL - Query Structure ReviewStructurally, each SQL query we have seen in this lesson are similar. Each start with a clause telling SQL which operation to perform and the remaining lines provide more detailed information as to how we want SQL to go about performing each SQL Command.

http://www.tizag.com/sqlTutorial/sqldelete.php

http://www.tizag.com/sqlTutorial/sqldelete.php

http://www.tizag.com/sqlTutorial/sqlupdate.php

http://www.tizag.com/sqlTutorial/sqlinsert.php

SQL - SelectSQL SELECT may be the most commonly used command by SQL programmers. It is used to extract data from databases and to present data in a user-friendly table called the result set.


SQL Select Query Template:SELECT table_column1, table_column2, table_column3 FROM my_table;

Select queries require two essential parts. The first part is the "WHAT", which determines what we want SQL to go and fetch. The second part of any SELECT command is the "FROM WHERE". It identifies where to fetch the data from, which may be from a SQL table, a SQL view, or some other SQL data object

Now we would like SQL to go and fetch some data for us from the orders table that was created in the previous lesson. How do we translate this request into SQL code so that the database application does all the work for us? Simple! We just need to tell SQL what we want to select and from where to select the data, by following the schema outlined below.

SQL Select Query Code:USE mydatabase;

SELECT id, customer, day_of_order, product, quantityFROM orders;

SQL Orders Table Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Pen4Below, we will manipulate the result output by rearranging the list of table column names inside of the SELECT statement.

SQL Select Query: Rearranged:


USE mydatabase;

SELECT day_of_order, customer, product, quantityFROM orders;

SQL Orders Table Results:day_of_ordercustomerproductquantity2008-08-01 00:00:00.000TizagPen4By rearranging the table column list inside the SELECT statement, we altered the appearance of the result set. Also, by not including the id column in the list of table columns, SQL did not fetch any column data for this column because we didn't ask SQL to do so.

SQL - Select All (*)"SELECT (*)" is a shortcut that can be used to select all table columns rather than listing each of them by name. Unfortunately, going this route doesn't allow for you to alter the presentation of the results.

SQL Select All Query:USE mydatabase;

SELECT *FROM orders;

SQL Orders Table Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Pen4

SQL - Selecting Data

The (*) query statement should be used with caution. Using this against our little tutorial database will surely do no harm, but using this query against an extremely large database may not be the best practice. Large databases may have web services or applications attached to them, so frequently updating and accessing large quantities data may temporarily lock a table for fractions of a second or more. If this disruption happens to occur just as some piece of data is being updated, you may experience data corruption.

Taking every precaution to avoid data corruption is in your best interest as a new SQL programmer. Corrupted data may be lost and never recovered, and it can lead to even more corruption inside a database. The best habits are to be as precise as possible, and in the case of select statements, this often means selecting minimal amounts of data when possible.

At this point, you should feel comfortable with SELECT and how to look into your database and see actual data rows residing inside of tables. This knowledge will prove invaluable as your SQL skills develop beyond the basics and as you begin to tackle larger, more advanced SQL projects.

SQL - WhereThe WHERE clause sets a conditional statement, and it can be used with any type of SQL query. As the select query executes, SQL processes one row at a time. Each time the conditional statement is met (returns true), a row is returned as a result. SQL WHERE is essentially, a filtering mechanism for SQL queries and is a tremendous asset to any aspiring SQL developer.


SQL Where Query:USE mydatabase;

SELECT *FROM ordersWHERE customer = 'Tizag'

As we take a look at the results, notice how only the rows that meet the criteria (where the customer column value is Tizag) are returned. In this example, we are using the WHERE clause to filter out rows and only selecting data that meets the conditional statement.

SQL Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Pen42Tizag2008-08-01 00:00:00.000Stapler15Tizag2008-07-25 00:00:00.00019" LCD Screen36Tizag2008-07-25 00:00:00.000


HP Printer2Conditional statements are not unique to SQL, and neither are operators. Operators are symbols such as (=) or (<), and they are seen inside of conditional statements and expressions in SQL and other programming languages. While we're not going to dive into much detail about the different kinds of operators yet, it is a good idea to be familiar with them and be able to recognize them inside of conditional statements as we look over the next few examples.

SQL - Where QueriesWith the WHERE clause on our tool belts, we can be more creative when querying for table rows. For instance, there may come a time where we would like to take a look at all the orders placed after a certain date.

SQL Where Date Query:USE mydatabase;

SELECT *FROM ordersWHERE day_of_order > '7/31/08'

This conditional statement will return only the orders that have made it into the table since the end of July, filtering out any orders in the table made prior to July 31st.

SQL Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Pen42Tizag2008-08-01 00:00:00.000Stapler13A+Maintenance2008-08-16 00:00:00.000Hanging Files

124Gerald Garner2008-08-15 00:00:00.00019" LCD Screen3Notice how the date value is formatted inside the conditional statement. We passed a value formatted MM/DD/YY, and we've completely neglected the hours, minutes, and seconds values, yet SQL is intelligent enough to understand this. Therefore, our query is successfully executed.

SQL - Where with Multiple ConditionalsA WHERE statement can accept multiple conditional statements. What this means is that we are able to select rows meeting two different conditions at the same time.

Perhaps the easiest way to go about this is to add another condition to the previous example, where we retrieved only the orders placed after July 31st. We can take this example one step further and link two conditional statements together with "AND".

SQL Where And:USE mydatabase;

SELECT *FROM ordersWHERE day_of_order > '7/31/08'AND customer = 'Tizag'

At this point, we have sent SQL two conditional statements with a single WHERE clause, essentially applying two filters to the expected result set.

SQL Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Pen42Tizag2008-08-01 00:00:00.000

Stapler1By applying the AND clause, SQL has now been asked to return only rows that meet both conditional statements. In this case, we would like to return all orders that were made before July 31st and made by a specific company - which is, in this case, Tizag. We have more examples of SQL AND/OR. Just follow the link.

http://www.tizag.com/sqlTutorial/sqlandor.php

SQL - AsSQL AS temporarily assigns a table column a new name. This grants the SQL developer the ability to make adjustments to the presentation of query results and allow the developer to label results more accurately without permanently renaming table columns.


SQL Select As Code:USE mydatabase;

SELECT day_of_order AS "Date", customer As "Client", product, quantityFROM orders;

SQL Orders Table Results:DateClientproductquantity2008-08-01 00:00:00.000TizagPen4SQL AS allows us to use any name at the presentation level and helps the developer better describe a column in the result set.

SQL Select Arithmetic Query:USE mydatabase;

SELECT (5 + 12) AS "5 plus 12 is"

SQL Arithmetic Results:5 plus 12 is17


SQL - OperatorsSQL operators are found in just about every SQL query. Operators are the mathematical and equality symbols used to compare, evaluate, or calculate values. Equality operators include the (<), (>), and (=) symbols, which are used to compare one value against another. Each of these characters have special meaning, and when SQL comes across them, they help tell SQL how to evaluate an expression or conditional statement. Most operators will appear inside of conditional statements in the WHERE clause of SQL Commands.


Operators come in three flavors: mathematical, logical, and equality. Mathematical operators add, subtract, multiply, and divide numbers. Logical operators include AND and OR. Take note of the following tables for future reference.

SQL operators are generally found inside of queries-- more specifically, in the conditional statements of the WHERE clause.

SQL Equality Operator Query:USE mydatabase;

SELECT customer,day_of_orderFROM ordersWHERE day_of_order > '7/31/08'

Sql Equality Operator:customerday_of_orderTizag2008-08-01 00:00:00.000Tizag2008-08-01 00:00:00.000In this case, we've used the equality operator greater than (>) to return orders from the orders table with a date greater than '7/31/08'.

SQL - Equality Operator TableEquality involves comparing two values. To do so requires the use of the (<), (>), or (=) special characters. Does X = Y? Is Y < X? These are both questions that can be answered using a SQL Equality Operator expression.


SQL Equality Operators:OperatorExampleDefinedResult=, IS5 = 55 equal to 5?True!=, IS NOT7 != 27 IS NOT (!=) equal to 2?True<7 < 47 less than 4?False>7 > 4greater than 4?True<=7 <= 11Is 7 less than or equal to 11?True>=7 >= 11Is 7 greater than or equal to 11?False

SQL - Mathematical OperatorsSQL mathematical operations are performed using mathematical operators (+, -, *, /, and %). We can use SQL like a calculator to get a feel for how these operators work.

SQL Mathematical Operators:SELECT 15 + 4, --Addition15 - 4, --Subtraction15 * 4, --Multiplication15 / 5, -- Division15 % 4; --Modulus

SQL Results:

AdditionSubtractionMultiplicationDivisionModulus19116033Modulus may be the only unfamiliar term on the chart. Modulus performs division, dividing the first digit by the second digit, but instead of returning a quotient, a "remainder" value is returned instead.

Modulus Example:USE mydatabase;

SELECT (5 / 2) -- = 2.5SELECT (5 % 2) -- = 1 is the value that will be returned

SQL - Logical OperatorsThese operators provide you with a way to specify exactly what you want SQL to go and fetch, and you may be as specific as you'd like! We'll discuss these a little later on and provide some real world scenarios as well.

We cover these operators thoroughly in the SQL AND/OR lesson.

AND - Compares/Associates two values or expressions OR - Compares/Associates two values or expressions

http://www.tizag.com/sqlTutorial/sqlandor.php

SQL - CreateSQL CREATE is the command used to create data objects, including everything from new databases and tables to views and stored procedures. In this lesson, we will be taking a closer look at how table creation is executed in the SQL world and offer some examples of the different types of data a SQL table can hold, such as dates, number values, and texts.


To accomplish this, it is best to first take a look at the entire CREATE TABLE query and then review each line individually.

SQL Create Table Code:USE mydatabase;

CREATE TABLE inventory( id INT IDENTITY(1,1) PRIMARY KEY, product VARCHAR(50) UNIQUE, quantity INT, price DECIMAL(18,2));

Line 1 identifies the scope of the query specifying a target database for query execution (USE mydatabase) and we've seen it before so let's skip ahead to the next line, line 2 (CREATE TABLE inventory). This line informs SQL of the plan to create a new table using the CREATE clause and specifies the name of the new table (inventory). In this case, we plan on creating an inventory table to maintain a current inventory of store items for an imaginary e-commerce web site.

SQL Create Line 3: id INT IDENTITY(1,1) PRIMARY KEY,

Line 3 should appear more foreign as there is a lot of information embedded in this line, but it is not as hard as it seems. This is the first line that declares how to set up the first table column inside the new inventorytable.

id = The name of this new table column. INT = The data type. INT is short for integer.

The first word, id, is the name of this new column and the second word declares the data type, INT (integers). SQL will now expect this table column to house only integer data type values.

IDENTITY (1,1) = The id column will be an identity column.

The next phrase, IDENTITY (1, 1) is a very special attribute and when a table column is marked as an


identity column, the column essentially turns into an automated counter. As new rows are inserted into the table, this column value will automatically increment (count up). The parameters (1,1) tell SQL which number to start counting from and by how many to increment each value. In this case, we'll start with 1, and increment by 1 each time a new row is inserted into our database. For example, the first INSERT command run against the inventory table will have an id value of 1, and each consecutive row inserted thereafter will increment by one (1, 2, 3, 4 ... etc). This identity table column is essentially counting each inserted row and also ensuring that we have a unique identifier value. This is important since this column has also been identified as a primary key (see below).

PRIMARY KEY = This places a restraint on this column (no duplicate values)

Bringing up the tail-end of Line 3 is reserved for specifying any unique attributes to associate with this table column. In this case, we have told SQL that this column will act as the PRIMARY KEY for the inventorytable. Declaring this column as the PRIMARY KEY places a restraint on this column meaning no duplicate values may exist in this column and SQL will throw an error message if an attempt is made to enter duplicate data. Since this row is set to automatically increment each time a new record is added, we know that this column will always be a unique value.

SQL Create Line 4: product VARCHAR(50) UNIQUE,

Line 4 specifies the name and type of the second column in the inventory table. Product stands for the inventory table product name and this column is set as a VARCHAR(50), which means it will be able to handle numbers, letters, and special characters as values. In other words, "Any words, numbers, or special characters can be placed into this column value, with a 50 character limit."

The UNIQUE attribute tells SQL that this table column must be a UNIQUE value at all times. This restraint will stop us from accidentally inserting duplicate records for the same product, which will serve as an aid to us to help maintain data integrity.

SQL Create Line 5,6: quantity INT, price DECIMAL(18,2)

Now that you are more familiar with the structure of this query, lines 5 and 6 should look less like a foreign language and more like SQL code. These lines are creating two more table columns: quantity and priceSince the price column will be dealing with decimals, we have set this column to a DECIMAL data type to handle decimals (sometimes called floating point numbers).

And there you have it, here's another look at the query:

SQL Create Table Query:USE mydatabase;

CREATE TABLE inventory( id INT IDENTITY(1,1) PRIMARY KEY, product VARCHAR(50) UNIQUE, quantity INT, price DECIMAL(18,2));

This SQL command will create a new, empty table called inventory, where we will begin to capture store inventory data to keep track of the price and current stock of items for our make believe online store.

At this point, this table has been created but remains empty, containing no data. Let's go ahead and add some records into this table so that we can then use this table to further our learning of SQL. Since we are now already familiar with the INSERT command, we can run the following commands all at once, so feel free to copy and paste this code into the query window and execute.

SQL Insert Into:USE mydatabase;

INSERT INTO inventoryVALUES('19" LCD Screen','25','179.99');INSERT INTO inventoryVALUES('HP Printer','9','89.99');INSERT INTO inventoryVALUES('Pen','78','0.99');INSERT INTO inventoryVALUES('Stapler','3','7.99');INSERT INTO inventoryVALUES('Hanging Files','33','14.99');INSERT INTO inventoryVALUES('Laptop','16','499.99');

Successful execution of the above query should yield messages indicating that the queries have run successfully.

SQL Results:(1 row(s) affected)(1 row(s) affected)(1 row(s) affected)(1 row(s) affected)(1 row(s) affected)(1 row(s) affected)We can double-check the results by running a SELECT (*) query and doing so will retrieve all records SQL has stored inside the inventory table.

SQL Code:USE mydatabase;

SELECT *FROM inventory;

SQL Results:idproductquantityprice119" LCD Screen25179.992HP Printer989.993Pen780.994Stapler37.995Hanging Files3314.996Laptop16499.99In creating this new table with data that relates to data inside the orders table, you have unknowingly created a relational database. We can now take a list of items ordered by our customers and verify that these items are in stock as purchases continue to flow in, so long as we maintain an up-to-date inventory table.

This is terrific news as you are now well on your way to take your SQL programming skills to the next level!

SQL - InsertTo use the INSERT command, we must first have an understanding of where we would like to insert data and what types of data we want to insert. Do we plan on inserting numbers? Strings? Files? Let's return to the orders table we created in an earlier lesson.


SQL tables store data in rows, one row after another. The INSERT command is the command used to insert new data (a new row) into a table by specifying a list of values to be inserted into each table column. The arrangement of values is important, and how they are arranged in the code corresponds to how the data values will be arranged in the the SQL table.

id - (identity, integer) customer - (customer name, character string) day_of_order - (date value) product - (name of product, character string) quantity - (quantity, integer)

Looking at the column names alone will give you an idea of what type of data each column is expected to hold. The quantity column, for example, is expecting a number or integer of some sort and the day_of_ordercolumn is expecting a date value to be inserted.

SQL Insert Query:USE mydatabase;

INSERT INTO orders (customer,day_of_order,product, quantity)VALUES('Tizag','8/1/08','Stapler',1);

SQL Insert Results:(1 row(s) affected)You may notice that the id column has been left out of the query statement. The reason behind this is that when we created the orders table, we gave the id column a unique attribute called identity. SQL handles identity columns automatically for us and therefore, we do not need to manually insert data into this column.

The first value Tizag corresponds with the customer table column. This ensures SQL will insert this value into the corresponding table column.

Now when we run the SELECT (*) query, SQL should return two rows with our statement instead of only a single row.

Verification Query:


USE mydatabase;

SELECT *FROM orders;

SQL Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Pen42Tizag2008-08-01 00:00:00.000Stapler1

SQL - Inserting ValuesAs a shortcut, you may omit the table columns entirely and only supply the values in the INSERT statement:

SQL Insert Shortcut:USE mydatabase;

INSERT INTO orders VALUES('A+Maintenance','8/16/08','Hanging Files',12);

Again, we can skip the id column because SQL is able to identify that this column is an identity column and handle it accordingly.

SQL Results:idcustomerday_of_orderproductquantity1Tizag

2008-08-01 00:00:00.000Pen42Tizag2008-08-01 00:00:00.000Stapler13A+Maintenance2008-08-16 00:00:00.000Hanging Files12Before moving on, let's add some more rows and execute some more INSERT queries. If you are using SQL Express, you should be able to copy the entire code section below and execute all the queries at once and then track the results with the verification query (SELECT * FROM orders).

SQL Inserts:USE myDatabase;

INSERT INTO ordersVALUES('Gerald Garner','8/15/08','19" LCD Screen',3)INSERT INTO orders VALUES('Tizag','7/25/08','19" LCD Screen',3);INSERT INTO orders VALUES('Tizag','7/25/08','HP Printer',2);

Final Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Pen42Tizag2008-08-01 00:00:00.000Stapler13A+Maintenance2008-08-16 00:00:00.000

Hanging Files124Gerald Garner2008-08-15 00:00:00.00019" LCD Screen35Tizag2008-07-25 00:00:00.00019" LCD Screen36Tizag2008-07-25 00:00:00.000HP Printer2

SQL - AndSQL AND links together two or more conditional statements for increased filtering when running SQL commands. AND helps the developer query for very specific records while answering questions like, "I want to view all orders made by a certain customer AND made on a special date." There is no limit to the number of AND/OR conditions that can be applied to a query utilizing the WHERE clause. This makes it possible for the developer to be as precise as needed when querying for results.


SQL And Code:USE mydatabase;

SELECT *FROM ordersWHERE customer = 'Tizag'AND day_of_order = '08/01/08'AND product = 'Pen';

SQL Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Pen4This example illustrates how SQL AND combines multiple conditional statements (3 total now) into a single condition with multiple circumstances (filters). Each filter removes rows from the result set that doesn't meet the condition.

SQL - OrSQL OR also applies logic to help filter results. The difference is that instead of linking together conditional statements, an OR condition just asks SQL to look for 2 separate conditions within the same query and return any records/rows matching either of the conditions.

SQL Or Code:USE mydatabase;


SELECT *FROM ordersWHERE product = 'Pen'OR product = '19" LCD Screen';

SQL Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Pen44Gerald Garner2008-08-15 00:00:00.00019" LCD Screen35Tizag2008-07-25 00:00:00.00019" LCD Screen3The first record returned matches the first condition since the product = 'Pen'. The two records after that match the other condition; the product in each of those orders is the '19" LCD Screen'. This type of logic allows the developer to filter results based on one or more conditions.

SQL AND allows the developer to query for more specific data by linking together conditional statements. On the other end of the spectrum, SQL OR creates a new, independent condition not linked to any other conditional statement.

SQL - And / Or CombinationConditional statements can be grouped together using parentheses (). Doing so links together conditions and provides robust solutions for data querying.

SQL And/Or Code:USE mydatabase;

SELECT *FROM orders

WHERE (quantity > 2 AND customer = 'Tizag')OR (quantity > 0 AND customer = 'Gerald Garner')

By encapsulating the first two conditions (quantity > 2 AND customer = 'Tizag') SQL now treats this as a single condition, and the same is true for the next line. These two conditions have been linked together with the OR operator, creating very unique behavior.

SQL is now looking for rows where the customer is Tizag AND the quantity is more than 2, and ALSO looking for rows where the customer is Gerald Garner AND the quantity is greater than 0. All rows meeting either condition will be returned as demonstrated in our results below.

SQL Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Pen44Gerald Garner2008-08-15 00:00:00.00019" LCD Screen35Tizag2008-07-25 00:00:00.00019" LCD Screen3

SQL - BetweenBETWEEN is a conditional statement found in the WHERE clause. It is used to query for table rows that meet a condition falling between a specified range of numeric values. It would be used to answer questions like, "How many orders did we receive BETWEEN July 20th and August 5th?"

SQL Select Between:USE mydatabase;

SELECT *FROM ordersWHERE day_of_order BETWEEN '7/20/08' AND '8/05/08';

SQL Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Pen42Tizag2008-08-01 00:00:00.000Stapler15Tizag2008-07-25 00:00:00.00019" LCD Screen36Tizag2008-07-25 00:00:00.000HP Printer2BETWEEN essentially combines two conditional statements into one and simplifies the querying process for you. To understand exactly what we mean, we could create another query without using the BETWEENcondition and still come up with the same results, (using AND instead).

SQL Select Between:USE mydatabase;

SELECT *FROM ordersWHERE day_of_order >= '7/20/08'AND day_of_order <= '8/05/08';

SQL Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Pen42Tizag2008-08-01 00:00:00.000Stapler15Tizag2008-07-25 00:00:00.00019" LCD Screen36Tizag2008-07-25 00:00:00.000HP Printer2As you can see from comparing the results of these two queries, we are able to retrieve the same data, but you may find BETWEEN easier to use and less cumbersome than writing two different conditional statements. In the end, the preference is really up to the individual writing the SQL Code.

SQL - Order ByORDER BY is the SQL command used to sort rows as they are returned from a SELECT query. SQL order by command may be added to the end of any select query and it requires at least one table column to be specified in order for SQL to sort the results.


SQL Order by query:USE mydatabase;

SELECT *FROM ordersWHERE customer = 'Tizag'ORDER BY day_of_order;

Executing this query should offer a list of orders made by Tizag and you may noticed that the result set has now been sorted (low to high) according to the date value. In other words, the oldest order to the newest order.

SQL Results:idcustomerday_of_orderproductquantity5Tizag2008-07-25 00:00:00.00019" LCD Screen36Tizag2008-07-25 00:00:00.000HP Printer21Tizag2008-08-01 00:00:00.000Pen42Tizag2008-08-01 00:00:00.000Stapler


1

SQL - Ascending DescendingThe default sort order for ORDER BY is an ascending list, [a - z] for characters or [0 - 9] for numbers. As an alternative to the default sorting for our results, which is ASCENDING (ASC), we can instead tell SQL to order the table columns in a DESCENDING (DESC) fashion [z-a].

SQL Order by Descending:USE mydatabase;

SELECT *FROM ordersWHERE customer = 'Tizag'ORDER BY day_of_order DESC

SQL Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Pen42Tizag2008-08-01 00:00:00.000Stapler15Tizag2008-07-25 00:00:00.00019" LCD Screen36Tizag2008-07-25 00:00:00.000HP Printer2If you compare these results to the results above, you should notice that we've pulled the same information but it is now arranged in a reverse (descending) order.

SQL - Sorting on Multiple ColumnsResults may be sorted on more than one column by listing multiple column names in the ORDER BY clause, similar to how we would list column names in each SELECT statement.

SQL Order by Multiple columns:USE mydatabase;

SELECT *FROM ordersORDER BY customer, day_of_order;

This query should alphabetize by customer, grouping together orders made by the same customer and then by the purchase date. SQL sorts according to how the column names are listed in the ORDER BY clause.

SQL Results:idcustomerday_of_orderproductquantity3A+Maintenance2008-08-16 00:00:00.000Hanging Files124Gerald Garner2008-08-15 00:00:00.00019" LCD Screen35Tizag2008-07-25 00:00:00.00019" LCD Screen36Tizag2008-07-25 00:00:00.000HP Printer21Tizag2008-08-01 00:00:00.000

Pen42Tizag2008-08-01 00:00:00.000Stapler1

SQL - UpdateSQL UPDATE is the command used to update existing table rows with new data values. UPDATE is a very powerful command in the SQL world. It has the ability to update every single row in a database with the execution of only a single query. Due to UPDATE's supreme authority, it is in your best interest to always include a WHERE clause when working with UPDATE query statements. That way, you will not accidentally update more rows than you intend to.


Execute the following UPDATE command to update the customer orders table. Since we've provided a WHERE condition with this update command, this update will only modify rows that match the condition and in this case it happens to be order number 1 made by Tizag. This update should increase the quantity from 4 Pens to 6 Pens for Tizag's first order.

SQL Update Query:USE mydatabase;

UPDATE ordersSET quantity = '6'WHERE id = '1'

SQL Results:(1 row(s) affected)Let's verify our results by selecting this row from the orders table.

SQL Verification Query:USE mydatabase;

SELECT *FROM ordersWHERE id = '1'

SQL Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000


Pen6The orders table now indicates that the customer Tizag will be ordering 6 Pens instead of 4. If the WHEREcondition is removed from this statement, SQL would modify every row with the new quantity value of instead of just the single row that meets the condition of id = "1". SQL UPDATE replaces data, much like overwriting a previously saved file on a computer hard drive. Once you click "Save," the old file is lost and replaced with the new file. Once an UPDATE command has been executed, the old data values are lost, being overwritten by the new value.

SQL - Update Incrementing a ValueIn the previous example, an order quantity was updated from 4 to 6. Say what we really wanted to do was not necessarily change it to 6, but to add 2 to the original order quantity. Updating the order quantity from 4 to 6 might have gotten the job done in that scenario, but that solution doesn't scale well. In the long run, we wouldn't get very much "bang for our buck," as they say.

So, perhaps a better way to tackle the same problem would be to increment the existing value (add 2) rather than updating with a single, static value. So, instead of setting the quantity table column to a specific value of 6, we can send it the current table column value directly and then add 2 to that already existing value.

SQL Update Code:USE mydatabase;

UPDATE ordersSET quantity = (quantity + 2)WHERE id = '1'

SQL Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Pen8Executing this update statement instead of the first update query is a huge timesaver. We no longer need to know the quantity of the order beforehand and we can add or subtract values from it in its current state. All we need to know is that we need to add 2 more to the quantity column to update the order correctly. This query is also more scalable, meaning we can update many rows at once. We will do so in the next example.

SQL - Update Multiple RowsAs mentioned earlier, removing the WHERE clause from any UPDATE command is generally not a good idea since doing so will result in SQL updating every row in the table. However, since the intention of this next example is to update multiple rows, let's go ahead and remove the WHERE clause from the above example.

SQL Update Multiple Rows:USE mydatabase;

UPDATE ordersSET quantity = (quantity + 2)

SQL Results:(6 row(s) affected)

SQL Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Pen102Tizag2008-08-01 00:00:00.000Stapler33A+Maintenance2008-08-16 00:00:00.000Hanging Files144Gerald Garner2008-08-15 00:00:00.00019" LCD Screen55

Tizag2008-07-25 00:00:00.00019" LCD Screen56Tizag2008-07-25 00:00:00.000HP Printer4

SQL Update Multiple ValuesSQL UPDATE can also be utilized to change multiple column values at once. Once again, let's update the same order id (1) changing the quantity of products ordered. But let's also take it another step further, by changing the quantity only when the products are Hanging Files.

SQL Update Multiple Values:USE mydatabase;

UPDATE ordersSET quantity = '11',Product = 'Hanging Files'WHERE id = '1'

SQL Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Hanging Files11The results show that we have successfully updated an order (order id 1). Notice that after the SET keyword, the column and value sets are listed with each column/value pair being separated with a comma (,).

SQL - AlterSQL ALTER is the command used to add, edit, and modify data objects like tables, databases, and views. ALTER is the command responsible for making table column adjustments or renaming table columns. New table columns can also be added and dropped from existing SQL tables.


SQL Add:USE mydatabase;

ALTER TABLE ordersADD discount VARCHAR(10);

SQL Results:idcustomerday_of_orderproductquantitydiscount1Tizag2008-08-01 00:00:00.000Pen8NULL2Tizag2008-08-01 00:00:00.000Stapler3NULL3A+Maintenance2008-08-16 00:00:00.000Hanging Files14NULL4Gerald Garner2008-08-15 00:00:00.00019" LCD Screen5


NULL5Tizag2008-07-25 00:00:00.00019" LCD Screen5NULL6Tizag2008-07-25 00:00:00.000HP Printer4NULLAs you can see from the results panel, SQL has added an additional column, discount, to the orders table. Since this column was just created, it contains no data, and only NULL values have been returned.

SQL - Alter Table: Modify ColumnSQL table columns can be altered and changed using the MODIFY COLUMN command. This allows the developer the opportunity to mold table columns or adjust settings as needed.

SQL Modify Column:USE mydatabase;

ALTER TABLE ordersALTER COLUMN discount DECIMAL(18,2);

Above, we have modified the new discount table column changing the column data type from a varchardecimal table column. This example can be expanded to modify table columns as needed by the developer.

SQL - SQL Alter Table: DropThis column can be deleted using the SQL DROP command. Once this column has been dropped, however, the data stored inside of it will be lost forever. Proceed with caution!

SQL Drop Column Code:USE mydatabase;

ALTER TABLE ordersDROP COLUMN discount;

SQL - DistinctSQL SELECT DISTINCT is a very useful way to eliminate retrieving duplicate data reserved for very specific situations. To understand when to use the DISTINCT command, let's look at a real world example where this tool will certainly come in handy.


If you've been following along in the tutorial, we have created an orders table with some data inside that represents different orders made by some of our very loyal customers over a given time period. Let's pretend that we have just heard word from our preferred shipping agent that orders made in August require no shipping charges, and we now have to notify our customers. We do not want to send mailers to all of our customers, just the ones that have placed orders in August. Also, we want to avoid retrieving duplicate customers as our customers may have placed more than one order during the month of August.

We can write a very simple SQL query to extract this information from the orders table:

SQL Select Distinct:USE mydatabase;

SELECT DISTINCT customerFROM ordersWHERE day_of_order BETWEEN '7/31/08' AND '9/1/08';

SQL Results:customerA+MaintenanceGerald GarnerTizagRunning this query yields a list of all the customer's affected by our unexpected news from the shipping agency. With this list, we can now go about contacting each of these customers and informing them of the good news without worrying about contacting the same customer multiple times.


SQL - SubqueriesSubqueries are query statements tucked inside of query statements. Like the order of operations from your high school Algebra class, order of operations also come into play when you start to embed SQL commands inside of other SQL commands (subqueries). Let's take a look at a real world example involving the orderstable and figure out how to select only the most recent order(s) in our orders table.


To accomplish this, we are first going to introduce a built-in SQL function, MAX(). This function wraps around a table column and quickly returns the current highest (max) value for the specified column. We are going to use this function to return the current "highest", aka most recent date value in the orders table.

SQL Subquery Preview:USE mydatabase;

SELECT MAX(day_of_order)FROM orders

SQL Results:day_of_order2008-08-16 00:00:00.000Now we can throw this query into the WHERE clause of another SELECT query and obtain the results to our little dilemma.

SQL Select Subquery Code:USE mydatabase;

SELECT *FROM ordersWHERE day_of_order = (SELECT MAX(day_of_order) FROM orders)

:idcustomerday_of_orderproductquantity3A+Maintenance2008-08-16 00:00:00.000


Hanging Files14This query is a dynamic query as it pulls current information and will change if a new order is placed. Utilizing a subquery we were able to build a dynamic and robust solution for providing us with current order information.

SQL - JoinSQL JOIN joins together two tables on a matching table column, ultimately forming one single temporary table. The key word here is temporary. The tables themselves remain intact, and running a JOIN query does not in any way change the data or table structure. JOIN is another way to select specific data from two or more relational tables.


In order to perform a JOIN query, we need a few pieces of information: the name of the table and table column we want to join on and a condition to meet for the JOIN to happen. This should sound a little confusing as there is much going on in a JOIN query, so let's take a look at an example:

SQL Join Query Code:USE mydatabase;

SELECT *FROM ordersJOIN inventoryON orders.product = inventory.product;

SQL Join Results:idcustomerday_of_orderproductquantityidproductquantityprice1Tizag2008-08-01 00:00:00.000Hanging Files115Hanging Files3314.992Tizag2008-08-01 00:00:00.000Stapler3


4Stapler37.993A+Maintenance2008-08-16 00:00:00.000Hanging Files145Hanging Files3314.994Gerald Garner2008-08-15 00:00:00.00019" LCD Screen5119" LCD Screen25179.995Tizag2008-07-25 00:00:00.00019" LCD Screen5119" LCD Screen25179.996Tizag2008-07-25 00:00:00.000HP Printer42HP Printer989.99The line beginning with JOIN (Line 4) is where we tell SQL which table we would like to join. The next line (Line 5) is a different story. Here is where we have specified the condition to JOIN ON. In this case, both tables have identical product columns which makes them an ideal target for a join. Basically we are temporarily merging the tables connecting them where they match, the product column.

This type of join matches values from one table column with a corresponding value in another table and uses that match to merge the tables together. In our make-believe store world, this let's us join the inventory table with the orders table to show us all the items we currently have in stock for our customers and also the price

of each item.

Let's rework this query a bit and strip away a few of the table columns to make our results easier to read and understand. We will replace the (*) parameter with a list containing only the table columns we are interested in viewing.

SQL Join:USE mydatabase;

SELECT orders.customer,orders.day_of_order,orders.product,orders.quantity as number_ordered,inventory.quantity as number_instock,inventory.priceFROM ordersJOIN inventoryON orders.product = inventory.product

SQL Results:customerday_of_orderproductnumber_orderednumber_instockpriceTizag2008-08-01 00:00:00.000Hanging Files113314.99Tizag2008-08-01 00:00:00.000Stapler337.99A+Maintenance2008-08-16 00:00:00.000Hanging Files143314.99Gerald Garner2008-08-15 00:00:00.00019" LCD Screen

525179.99Tizag2008-07-25 00:00:00.00019" LCD Screen525179.99Tizag2008-07-25 00:00:00.000HP Printer4989.99Since we have one column in each table named the same thing (quantity), we used AS to modify how these columns would be named when our results were returned. These results should be more satisfying and easier to read now that we have removed some of the unnecessary columns.

SQL - Right JoinRIGHT JOIN is another method of JOIN we can use to join together tables, but its behavior is slightly different. We still need to join the tables together based on a conditional statement. The difference is that instead of returning ONLY rows where a join occurs, SQL will list EVERY row that exists on the right side, (The JOINED table).

SQL - Right Join:USE mydatabase;

SELECT *FROM ordersRIGHT JOIN inventoryON orders.product = inventory.product

SQL Results:idcustomerday_of_orderproductquantityidproductquantityprice4

Gerald Garner2008-08-15 00:00:00.00019" LCD Screen5119" LCD Screen25179.995Tizag2008-07-25 00:00:00.00019" LCD Screen5119" LCD Screen25179.996Tizag2008-07-25 00:00:00.000HP Printer42HP Printer989.99NULLNULLNULLNULLNULL3Pen780.992Tizag2008-08-01 00:00:00.000Stapler34Stapler37.991Tizag2008-08-01 00:00:00.000Hanging Files

115Hanging Files3314.993A+Maintenance2008-08-16 00:00:00.000Hanging Files145Hanging Files3314.99NULLNULLNULLNULLNULL6Laptop16499.99You should see a new row at the bottom of the results box with a bunch of NULL values. This is a result of the RIGHT JOIN and is the intended result from running the query. We end up with an extra row because inside of the inventory table, the Laptop item was not joined with a product from the orders table. This just means that we have not sold a laptop as of yet and it shouldn't be much a surprise since we already know from querying the orders table in previous lessons that there have been no laptop orders so far.

By specifying RIGHT JOIN, we have told SQL to join together the tables even if no matches are found in the conditional statement. All records that exist in the table on the right side of the conditional statement (ON orders.product = inventory.product) will be returned and NULL values will be placed on the left if no matches are found.

SQL - Left JoinSQL LEFT JOIN works exactly the same way as RIGHT JOIN except that they are opposites. NULL values will appear on the right instead of the left and all rows from the table on the left hand side of the conditional will be returned.

Unfortunately, we will not be able to show a very intuitive example of a LEFT JOIN because of how our tables are structured. The orders table should always have a matching inventory item and if not, that means we are in big trouble as we could be selling items we do not carry in inventory. For good measure, here's what a LEFT JOIN would look like:

SQL Left Join:USE mydatabase;

SELECT *FROM ordersLEFT JOIN inventoryON orders.product = inventory.product

SQL JOIN is intended to bring together data from two tables to form a single larger table, and often, it will paint a more detailed picture of what the data represents. By merging these two data sets, we were able to peer into our database and ensure that each item ordered so far is in stock and ready to be shipped to our customers.

SQL - InSQL IN is an operator used to pull data matching a list of values. A scenario where this proves useful would be if we wanted to retrieve customer data for two or more customers. We can use the IN operator to specify a list of customer names, and SQL will retrieve rows reflecting every customer in the list.


Inside the query statement itself, the word "IN" replaces the (=) operator after the WHERE declarative and slightly alters the meaning as well. Instead of listing a single value, you may list multiple values and SQL will retrieve the results for each value listed.

SQL In:USE mydatabase;

SELECT *FROM ordersWHERE customer IN ('Gerald Garner','A+Maintenance');

SQL Results:idcustomerday_of_orderproductquantity3A+Maintenance2008-08-16 00:00:00.000Hanging Files144Gerald Garner2008-08-15 00:00:00.00019" LCD Screen5The results provide a list of all customer orders made by each of the customer names we have listed inside the IN clause ('Gerald Garner','A+Maintenance'). This is a great way to query for all orders made by a handful of different customers as we can see everything these particular customers have ordered thus far.

The real power of this condition comes to life when used with a subquery that retrieves a list of values. Running any SELECT query returns results in list format. And as we mentioned just a few short moments ago, this list can then be passed as a list for the IN clause using a subquery.

Let's adjust the previous example to only retrieve only the products column, as opposed to retrieving all


columns (*).


SELECT productFROM ordersWHERE customer IN ('Gerald Garner','A+Maintenance');

Results:productHanging Files19" LCD ScreenOur results represent a query run to achieve a list of products sold to two of our customers. Now let's convert this query to a subquery and use this query as an input list to check the inventory table to see if we have any of these items in stock.


SELECT *FROM inventoryWHERE product in (SELECT product FROM orders WHERE customer IN ('Gerald Garner','A+Maintenance'));

SQL Results:idproductinventoryprice119" LCD Screen25179.995Hanging Files3314.99By specifying a sub query as our list of values we were able to take advantage of the relationship our tables have with each other and create a very dynamic query. This query saves us the time of scrolling through the

entire inventory table and checking the stock of each item purchased by any of our recent customers.

SQL - Not InSQL NOT IN, as you may have guessed, allows the developer to eliminate a list of specific values from the result set.

SQL Not In:USE mydatabase;

SELECT *FROM inventoryWHERE product NOT IN (SELECT product FROM orders WHERE customer IN ('Gerald Garner','A+Maintenance'));

SQL Results:idproductquantityprice2HP Printer989.993Pen780.994Stapler37.996Laptop16499.99

SQL - CaseSQL CASE is a very unique conditional statement providing if/then/else logic for any ordinary SQL command, such as SELECT or UPDATE. It then provides when-then-else functionality (WHEN this condition is met THEN do_this).


This functionality provides the developer the ability to manipulate the presentation of the data without actually updating or changing the data as it exists inside the SQL table.

SQL Select Case Code:USE mydatabase;

SELECT product, 'Status' = CASE WHEN quantity > 0 THEN 'in stock' ELSE 'out of stock' ENDFROM dbo.inventory;

SQL Results:productStatus19" LCD Screenin stockHP Printerin stockPenin stockStaplerin stockHanging Filesin stockLaptopin stockUsing the CASE command, we've successfully masked the actual value of the product inventory without actually altering any data. This would be a great way to implement some feature in an online catalog to allow users to check the status of items without disclosing the actual amount of inventory the store currently has in stock.

SQL - Case: Real World ExampleAs a store owner, there might be a time when you would like to offer sale prices for products. This is a


perfect opportunity to write a CASE query and alter the inventory sale prices at the presentation level rather than actually changing the price inside of the inventory table. CASE provides a way for the store owner to mask the data but still present it in a useful format.

Let's back up a second and pull a listing of our recent orders and join this with the inventory table so that the results contain both the quantity of items purchased and the price from the inventory table. To accomplish this we will need to first write a SQL JOIN query.

SQL Join Query:USE mydatabase;

SELECT dbo.orders.id, dbo.orders.customer, dbo.orders.quantity, dbo.inventory.product, dbo.inventory.priceFROM ordersJOIN inventoryON orders.product = inventory.product

In order to provide results that are much clearer, we've moved away from selecting every column with (*). Instead, we've listed each column that will be of use for the next few steps. Also, let's plan on offering a 25off sale on these items.

SQL Results:idcustomerquantityproductprice1Tizag11Hanging Files14.992Tizag3Stapler7.993A+Maintenance14Hanging Files14.994

Gerald Garner519" LCD Screen179.995Tizag519" LCD Screen179.996Tizag4HP Printer89.99Next we need to look at reducing the prices of the items according to our sale price. For the purpose of this exercise, let's offer a 25% discount on all our currently pending orders using a SQL CASE query.

SQL Select Case Code:USE mydatabase;

SELECT dbo.orders.id, dbo.orders.customer, dbo.orders.quantity, dbo.inventory.product, dbo.inventory.price, 'SALE_PRICE' = CASE WHEN price > 0 THEN (price * .75) ENDFROM orders JOIN inventoryON orders.product = inventory.product

Multiplying the current price by .75 reduces the price by approximately 25%, successfully applying the changes we would like to see but doing so without actually changing any data.

SQL Results:idcustomerquantityproductpriceSALE_PRICE1Tizag11Hanging Files14.99

11.24252Tizag3Stapler7.995.99253A+Maintenance14Hanging Files14.9911.24254Gerald Garner519" LCD Screen179.99134.99255Tizag519" LCD Screen179.99134.99256Tizag4HP Printer89.9967.4925The results speak for themselves as the records returned indicate a new table column with the calculated sales price now listed at the end of each row.

Since SQL CASE offers a conditional statement (price > 0), it wouldn't take much more effort to create some conditional statements based on how many products each customer had ordered and offer different discounts based on the volume of a customer order.

For instance, as a web-company, maybe we would like to offer an additional 10% discount to orders totaling more than $100. We could accomplish this in a very similar fashion.

SQL Results:USE mydatabase;

SELECT dbo.orders.id, dbo.orders.customer, dbo.orders.quantity,

dbo.inventory.product, dbo.inventory.price, 'SALE_PRICE' = CASE WHEN (orders.quantity * price) > 100 THEN (price * .65) ELSE (price * .75) ENDFROM orders JOIN inventoryON orders.product = inventory.product

:idcustomerquantityproductpriceSALE_PRICE1Tizag11Hanging Files14.999.74352Tizag3Stapler7.995.99253A+Maintenance14Hanging Files14.999.74354Gerald Garner519" LCD Screen116.9935134.99255Tizag519" LCD Screen179.99116.99356

Tizag4HP Printer89.9958.4935With this query, we have now successfully reduced all orders by 25% and also applied an additional 10%

discount to any order totaling over $100.00.

In each of the examples above, SQL CASE has been utilized to perform presentation level adjustments on data values and its versatility provides limitless results.

SQL - Group BySQL GROUP BY aggregates (consolidates and calculates) column values into a single record value. GROUP BY requires a list of table columns on which to run the calculations. At first, this behavior will resemble the SELECT DISTINCT command we toyed with earlier.


SQL Group By:USE mydatabase;

SELECT customerFROM ordersGROUP BY customer;

SQL Results:customerA+MaintenanceGerald GarnerTizagHere, SQL has consolidated like values and returned those that are unique. In this case, we have actually duplicated the behavior of SELECT DISTINCT, but you have also seen firsthand how GROUP BY accepts a table column as a list and consolidates like customer values.

To unleash the true power of GROUP BY, it is necessary to include at least one mathematical (aggregate) function, and to do so we will utilize the SUM() function to calculate how many total items have been purchased by each of our customers.


SELECT customer, SUM(quantity) AS "Total Items"FROM ordersGROUP BY customer;

SQL Results:customerTotal ItemsA+Maintenance14Gerald Garner5


Tizag23With the addition of the aggregate SUM() function, we've let SQL calculate how many products have been ordered by each customer and returned them for viewing with a single query statement.

Taking a look at another example, we can also figure out how many of each product was ordered with the use of a single query statement.


SELECT product, SUM(quantity) AS "Total Items"FROM ordersGROUP BY product;

SQL Results:productTotal Items19" LCD Screen10Hanging Files25HP Printer4Stapler3GROUP BY would also be a great way to calculate how much total cash of our customers has spent. Let's take a look at what that query may look like.


SELECT customer, SUM((orders.quantity * inventory.price)) AS "COST"FROM ordersJOIN inventoryON orders.product = inventory.productGROUP BY customer;

SQL Results:productCOSTA+Maintenance

209.86Gerals Garner899.95Tizag1448.77

SQL - Grouping By Multiple ColumnsLike the ORDER BY clause, GROUP BY can accept a list of table columns on which to group by.


SELECT day_of_order, product, SUM(quantity) as "Total"FROM ordersGROUP BY day_of_order,productORDER BY day_of_order;

SQL Results:day_of_orderproductTotal2008-07-25 00:00:00.00019" LCD Screen52008-07-25 00:00:00.000HP Printer42008-08-01 00:00:00.000Hanging Files112008-08-01 00:00:00.000Stapler32008-08-15 00:00:00.00019" LCD Screen52008-08-16 00:00:00.000Hanging Files14This query will group together and sum the total number of products purchased on any given date, regardless of what customer has purchased the item. It's a very useful query to keep in mind.

SQL - HavingThe SQL HAVING clause is "like a WHERE clause for aggregated data." It's used with conditional statements, just like WHERE, to filter results. One thing to note is that any column name appearing in the HAVING clause must also appear in the GROUP BY clause.

SQL Having:USE mydatabase;

SELECT day_of_order, product, SUM(quantity) as "Total"FROM ordersGROUP BY day_of_order,product,quantityHAVING quantity > 7ORDER BY day_of_order;

SQL Results:day_of_orderproductTotal2008-08-01 00:00:00.000Hanging Files112008-08-16 00:00:00.000Hanging Files14The quantity column is now considered aggregated in SQL terms, because its values have been summed together using the SUM() function. In the example above, HAVING acts as the WHERE clause for aggregate values, filtering out results that do not meet the condition (quantity > 7).

SQL - ViewsSQL VIEWS are data objects, and like SQL Tables, they can be queried, updated, and dropped. A SQL VIEW is a virtual table containing columns and rows except that the data contained inside a view is generated dynamically from SQL tables and does not physically exist inside the view itself.


SQL Create View Code:CREATE VIEW virtualInventoryAS SELECT * FROM inventory;

With a successful execution of this query, we have now created a view data object of the inventory table. The virtualInventory view is considered a data object (like a table) and is now accessible to us the developer. Views can be queried exactly like any other SQL table.

SQL View Code:USE mydatabase;

SELECT *FROM virtualInventory;

SQL Results:idproductquantityprice119" LCD Screen25179.992HP Printer989.993Pen780.994Stapler3


7.995Hanging Files3314.996Laptop16499.99Even though a SQL VIEW is treated like a data object in SQL, no data is actually stored inside of the view itself. The view is essentially a dynamic SELECT query, and if any changes are made to the originating table(s), these changes will be reflected in the SQL VIEW automatically.


UPDATE inventorySET price = '1.29'WHERE product = 'Pen';

Execute the following query to verify the results:

SQL Verification Query Code:USE mydatabase;

SELECT *FROM virtualInventoryWHERE product = 'Pen';

SQL Results:idproductquantityprice3Pen781.29

SQL - Drop ViewViews can also be removed by using the DROP VIEW command.

SQL Drop View:USE mydatabase;

DROP VIEW virtualInventory;

SQL - DatesDate values are stored in date table columns in the form of a timestamp. A SQL timestamp is a record containing date/time data, such as the month, day, year, hour, and minutes/seconds. It's not much different from the standard date format.


Date Columns:Column TypeFormattimeHH:MM:SSdateYYYY-MM-DDdatetimeYYYY-MM-DD HH:MM:SSDate values are stored in the form of a timestamp, and SQL offers a built-in function called GETDATE() that returns the current date in the form of a SQL timestamp.

SQL SELECT GETDATE():USE mydatabase;

SELECT GETDATE();

Timestamp Result:2004-06-22 10:33:11.840SQL expects dates to be formatted as above but does offer some flexibility when working with dates inside query statements. For instance, date values do not necessarily need to contain the hour, minutes, and seconds values. SQL also accepts most traditional date formats such as "MM/DD/YY" (ex: "01/01/06").

Using a built in function, ISDATE() we can do some testing on date values to see if they meet the formatting requirements.


SELECT ISDATE('8/24/08') AS "MM/DD/YY",ISDATE('2004-12-01') AS "YYYY/MM/DD";


ISDATE() returns a 1 or a 0 indicating a true or false result. In this case, both formats are acceptable date formats as a 1 value was returned.

SQL - Month(), Day(), Year()The Month(), Day() and Year() functions all extract corresponding values from a given date.

SQL Year():USE mydatabase;

SELECT YEAR(GETDATE()) as "Year";SELECT YEAR('8/14/04') as "Year";

SQL Month():USE mydatabase;

SELECT MONTH(GETDATE()) as "Month";SELECT MONTH('8/14/04') as "Month";

SQL Day():USE mydatabase;

SELECT DAY(GETDATE()) as "Day";SELECT DAY('8/14/04') as "Day";

Understanding timestamps and extracting pieces of dates is the first step in being able to perform date calculations and work more in-depth with SQL Dates.

SQL - DatepartDATEPART() is a SQL function used to extract all kinds of date information from timestamps, and it is a function that is unique to Microsoft's SQL Server Application.


SQL Datepart:USE mydatabase;

SELECT DATEPART(year, '2007-06-01') AS "Year";

SQL Results:Year2007DATEPART() requires 2 parameters separated by a comma (,). The first parameter specifies what type of date data will be extracted, and the second parameter is a timestamp value.

SQL Datepart:USE mydatabase;

SELECT DATEPART(year, '2007-06-01') AS "Year", DATEPART(month, '2007-06-01') AS "Month", DATEPART(day, '2007-06-01') AS "Day", DATEPART(dayofyear, '2007-06-01') AS "DayofYear", DATEPART(weekday, '2007-06-01') AS "Weekday";

SQL Results:YearMonthDayDayofYearWeekday2007611526

Datepart Abbreviation Chart:DatePart


AbbreviationExampleyearyy, yyyyDATEPART(yy, getdate())quarterqq, qDATEPART(qq, getdate())monthmm, mDATEPART(mm, getdate())dayofyeardy, yDATEPART(dy, getdate())daydd, dDATEPART(dd, getdate())weekwk, wwDATEPART(wk, getdate())weekdaydwDATEPART(dw, getdate())hourhhDATEPART(hh, getdate())minutemiDATEPART(mi, getdate())secondssDATEPART(ss, getdate())millisecondmsDATEPART(ms, getdate())

SQL - DateAdd()DATEADD() is the SQL function used to add and increment date values. Hours, minutes, months, and days can be added to any date value. In fact, dates can be added based on any type of date part discussed in the SQL DATEPART() lesson.



SELECT DATEADD(year, 1, getdate()) AS "+1 Year";

SQL Results:+1 Year2009-06-31 00:00:00.000This example shows how to use DATEADD() to take a specified date value and increment it by the 'year' date part. By replacing the middle parameter with a negative value, we can utilize the same DATEADD()function to subtract dates as well.


SELECT DATEADD(day,-1, '2006-06-01') AS "-1 Day";

SQL Results:-1 Day2006-05-31 00:00:00.000In each example, SQL is able to perform a calculation on each date value based on a timestamp, and after the calculation, a timestamp value returned. Also note that the date parameter can be based on another SQL function or the result of a subquery.


SELECT DATEADD(day,-30, (SELECT MAX(day_of_order) FROM orders)) AS "-30 Days";

SQL Results:


http://sqldatepart.php/

-30 Days2008-07-17 00:00:00.000Here we have now constructed a very useful, dynamic statement pulling the most current order (MAX) in the orders table, and we've been able to subtract one day from that value. While this information does not directly prove useful, if we take this query one step further and place this statement in a WHERE as a subquery, we should be more satisfied with the results.


SELECT * FROM ordersWHERE day_of_order >(SELECT DATEADD(day,-30, (SELECT MAX(day_of_order) FROM orders)) AS "-30 Days");

SQL Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Hanging Files112Tizag2008-08-01 00:00:00.000Stapler33A+Maintenance2008-08-16 00:00:00.000Hanging Files144Gerald Garner2008-08-15 00:00:00.00019" LCD Screen55Tizag2008-07-25 00:00:00.00019" LCD Screen5

6Tizag2008-07-25 00:00:00.000HP Printer4By placing this calculated date in the WHERE clause, we were able to pull all the records that have happened within 30 days of the most recent order (2008-07-17 00:00:00.000). We are able to query the orders table and request this information with a dynamic query that will yield different results as new orders are placed and time goes by.

SQL - Delete Command(s)In the SQL world, databases, rows, and columns all have one thing in common: once a DELETE statement has been executed successfully against them, the data they once contained is lost forever! Be very careful with these commands and be sure to properly backup all data before proceeding with any type of DELETEcommand(s).


SQL offers several ways to tackle data deletion. Here are the differences.

SQL Delete Commands:DELETE - Deletes any number of rows from a data object.DROP - Removes table columns, tables, and all data objects SQL applications.TRUNCATE - Empties out data without removing the object itself.

SQL - DeleteDELETE queries work much like UPDATE queries and like UPDATE, it is much advised to always use a WHERE condition when running any delete query or else you risk deleting too much data.

SQL Delete Query:USE mydatabase;

DELETE FROM ordersWHERE customer = 'A+Maintenance';

SQL Results:1 Row(s) affected

SQL - TruncateSQL TRUNCATE is the fastest way to remove all data from a SQL table, leaving nothing but an empty shell. You might choose to use this command when all the data inside of a table needs to be removed but you'd like the table column definitions to remain intact.

SQL Truncate Table Code:USE mydatabase;


TRUNCATE TABLE orders;

NOTE: Executing the command above will empty your table data and you will lose this data forever! If you plan on following along do not execute this query.

SQL - DropSQL DROP is another command that removes data from the data store. The DROP command must be performed on SQL objects including databases, tables, table columns, and SQL views. Dropping any of these objects removes them completely from your SQL application and all data contained in any of the data objects dropped are lost forever.

SQL Drop Examples:USE mydatabase;

DROP TABLE orders;DROP DATABASE mydatabase;DROP VIEW viewname;DROP INDEX orders.indexname;

-- FOR USE WITH ALTER COMMANDSDROP COLUMN column_nameDROP FOREIGN KEY (foreign_key_name)

The above example also includes the syntax to drop table columns and foreign keys. These items are outlined in the SQL ALTER lesson.

http://sqlalter.php/

SQL - UnionSQL UNION combines two separate SQL queries into one result set. A JOIN statement adds additional table columns to a result set (horizontally), UNION combines row results from one table with rows of another table (vertically).


In order to perform a UNION the columns of table 1 must match those of table 2. This rule ensures that the result set is consistent as rows are fetched by SQL.

For these next exercises we suggest creating two different tables that are identical in structure but contain unique rows of data. We challenge you to do this by reviewing the SQL Create queries and modifying them to create two brand new employee tables.

SQL Select Union Code:USE mydatabase;

SELECT * FROM employeesUNIONSELECT * FROM employees2;

SQL Table:IDLastnameFirstnameTitle1JohnsonDavidcrew2HivelyJessicacrew9HicksFreddycrew10HarrisJoelcrew11

http://www.tizag.com/sqlTutorial/sqlcreate.php


DavisJuliemanager101YazzowJimcrew102AndersonCraigcrew103CarlsonKevincrew104MainesBradcrewThe result is a complete listing of every employee from the two tables, perhaps representing a list of employees from two different departments.

The next example shows a more practical means of using a union clause. Here we will select all of our employees from both tables and join them with our invoices table to generate a complete list of sales from both stores on a given day.

SQL Code:SELECT employees.Lastname, employees.Firstname, invoices.Sale, invoices.PriceFROM employeesINNER JOIN invoicesON employees.id = invoices.EmployeeIDUNIONSELECT employees2.Lastname, employees2.Firstname, invoices.Sale, invoices.PriceFROM employees2INNER JOIN invoicesON employees2.id = invoices.EmployeeID;

SQL Table:LastnameFirstnameSalePriceJohnsonDavidHOT DOG1.99

HivelyJessicaLG SFT DRK1.49DavisJulieCK SLD3.99YazzowJimHOT DOG1.99CarlsonKevinLG SFT DRK1.49Here we combined a join query with the union clause to create one table.

SQL - Union AllUNION ALL selects all rows from each table and combines them into a single table. The difference between UNION and UNION ALL is that UNION ALL will not eliminate duplicate rows. Instead, it just pulls all rows from all tables fitting your query specifics and combines them into a table.

SQL Code:SELECT * FROM employeesUNION ALLSELECT * FROM employees2;

SQL Table:IDLastnameFirstnameTitle1JohnsonDavidcrew2HivelyJessicacrew9Hicks

Freddycrew10HarrisJoelcrew11DavisJuliemanager101YazzowJimcrew102AndersonCraigcrew103CarlsonKevincrew11DavisJuliemanager104MainesBradcrew

SQL Code:SELECT employees.Lastname, employees.Firstname, invoices.Sale, invoices.PriceFROM employeesINNER JOIN invoicesON employees.id = invoices.EmployeeIDUNION ALLSELECT employees2.Lastname, employees2.Firstname, invoices.Sale, invoices.PriceFROM employees2INNER JOIN invoicesON employees2.id = invoices.EmployeeID;

SQL Table:LastnameFirstnameSalePrice

JohnsonDavidHOT DOG1.99HivelyJessicaLG SFT DRK1.49DavisJulieCK SLD3.9911DavisJuliemanagerYazzowJimHOT DOG1.99CarlsonKevinLG SFT DRK1.4911DavisJuliemanager11DavisJuliemanager

SQL - Syntax - (Speaking SQL)Syntax, by definition, means the study of linguistic rules and patterns. Every programming language, including SQL, must follow a unique set of guidelines termed syntax. Punctuation, spaces, mathematical operators, and special characters have special meaning when used inside of SQL commands and query statements. For example, each and every SQL command will end with a semi colon (;).


Executing SQL commands that do not have proper syntax and formatting will result in a syntax error. Syntax errors might be the most common and first error messages new SQL developers will experience.

Let's now take a look at a very simple SQL command that will be used in just about every example contained in this tutorial from here on out.

Sample SQL Command:use mydatabase;

This command identifies a database as a target database, meaning that any SQL command or query executed after this line will be run against the identified database. In this case, the database 'mydatabase' will be the target database. This is a good way to prevent mistakes and avoid potential data loss and a good reason to include this command into each and every authored SQL query.

SQL - Syntax: Capitalization and SpacingIn some programming languages, capitalizing commands or excessive spacing may or may not cause syntax code errors and cause the command to fail. SQL syntax is very loose when it comes to capitalization and spacing, allowing a lot of room for the developer to decide on his/her own preference in regards to capitalization and spacing.

Let's rewrite the same SQL command from the previous example and take advantage of SQL's loose syntax characteristics.

Sample SQL Command:USE

mydatabase;

The example above, though it does look different due to the capitalization and spacing, will yield the same results as the first example.


SQL - Syntax: Building good habitsWhile coding in any language, it is important to develop good, consistent habits and maintain clean code. Clean code allows another SQL developer to step right in where you have left off without missing a beat. A developer with diligent coding habits will prevent many syntax errors before executing his/her scripts and also will be able to detect possibly syntax problems before they cause problems in a SQL Command.

Good habits include:

Consitency Clean and Concise Use of Comments (more on this later) Scalability

Coding in any language is as much of an art form as authoring best selling novels and stories. Take pride in doing so and always do your best to follow good coding habits.

SQL - Data TypesSQL data takes shape in several different forms, including character strings, numbers, file stores, and dates. SQL developers call the shots as to what types of data will be stored inside each and every table column when creating a SQL table. The developer must specify the column type of each new SQL table column.


Column types are synonymous with data types as the column type is what designates the type of data that will be stored inside the column. In other words, a SQL data type is a label and a guideline for SQL to understand what type of data is expected inside of each table column and this identifies how SQL will interact with the stored data. Below, we will give you an overview on the types of data that can be stored within a SQL table.

SQL - Numbers, Decimals, and Dates

Data Types: Integers - (3, -17) Point(Decimal) - (3.23415) Date - (2004-06-22 10:33:11.840)

Storing numbers and decimals allows the developer to collect statistical data and create reports based on the data contained inside the table. SQL can even perform mathematical calculations against numeric data, providing endless number-crunching abilities.

In SQL, decimals are often referred to as point or floating-point numbers. These data types are slightly different from the normal 'integer' data types.

For the most part, date values are treated like numbers and they can even be added together and subtracted offering the developer the option to add days, months, or years together to create new dates (more on this later). Additionally, specific data can be extracted from date values, allowing the developer to pull specific date information from a date value like only the month number, the year, or the day of the week.

SQL - Boolean Data ("TRUE" / "FALSE") ( 1 / 0 )

Boolean values are true/false types of data. A Boolean table column will contain either string values of "True" and "False" or the numeric equivalent representation, with 0 being false and 1 being true.


SQL - Character Strings

Character Strings: VARCHAR - ('Words or numbers') Text - ('Once upon a time...')

Strings range from a single word or character to large blocks of text including multiple paragraphs and unique symbols. Set the table column type to VARCHAR or Text in order to incorporate string data types into SQL tables.

SQL Server Table Column Types:bigintInteger value (-9,223,372,036,854,775,808 - 9,223,372,036,854,775,807)2^63intsmaller Integer value (-2,147,483,648) - (2,147,483,647)2^31smallintsmaller Integer value (-32,768) - (32,767)2^15tinyintsmaller Integer values 0 - 2552^8bitInteger data value (either 1 or 0 value)1 or 0decimalDecimal values from -10^38 - 10^3810^38numericDecimal values from -10^38 - 10^3810^38moneyMoney values (-922,337,203,685,477.5808) - (922,337,203,685,477.5807)2^63smallmoneySmaller Money Values (-214,748.3648) - (214,748.3647)2^31datetimeDate value (January 1, 1753) - (December 31, 9999)smalldatetimeSmaller Date Value (January 1, 1900) - (June 6, 2079)timestamp

Unique Number Value (updates when row is updated)charCharacter String Value (max 8,000 characters)varcharCharacter String Value maximum of 8,000 characters, unless otherwise noted)ncharCharacter String Value (max 4,000 characters)nvarcharCharacter String Value (max 4,000 characters)textCharacter String Value (max 2,147,483,647 characters)2^31ntextCharacter String Value (max 1,073,741,823 characters)2^30binaryBinary Value (max 8,000 bytes)varbinaryBinary Value (max 8,000 bytes)imageBinary Value (max 2,147,483,647 bytes)2^31uniqueidentifierGlobal Unique ID (GUID)

SQL - Defaults and Null ValuesNULL values are 'nothing' values. When a value is null, it means the value is empty and contains no value -- not even '0'. NULLs are unique data types that are usually the default setting for all table columns. When a SQL developer runs across a NULL value in a database, it is generally an indication that this value is either new or has not been modified.

The SQL developer may specify to allow or disallow the NULL values eliminating the possibility of running across 'empty' table columns when creating a SQL table. If the developer chooses not to allow NULL values he/she may specify a custom default value instead of the NULL (nothing) value. Primary Key table columns do not allow NULL values since this column's sole purpose is to be the unique identifier for a table column. Having a NULL unique identifier would be similar to having a car license plate that is blank.

By default, NULL values are allowed on all newly created table columns meaning a table column is allowed to be 'empty', except primary key columns. A NULL value is a special type of value that can be tested for by most programming languages including SQL and can provide the developer a means to 'test' and see if data exists or has been modified. As a new programmer you may not fully understand the benefits a NULL value can bring, but with experience, you will learn to hate/appreciate them.

SQL - ExpressionsSQL Expressions are the pieces of a SQL query that compare values against other values or perform arithmetic calculations. Expressions can be found inside of any SQL command usually in the form of a conditional statement. In the SQL world, conditional statements and expressions test or compare values against other values.


SQL - Boolean ExpressionsBoolean expressions return rows (results) when a single value is matched.

SQL Boolean Expression:USE mydatabase;

SELECT * FROM orders WHERE id = '1';

SQL Results:idcustomerday_of_orderproductquantity1Tizag2008-08-01 00:00:00.000Pen4

SQL - Numeric ExpressionNumeric Expressions return a single numeric value instead of an entire row and usually perform calculations.


SELECT 15 + 4;



SELECT (15 / 5) * 10;


SELECT ((5+5) * (5+5));

Each of the examples above returns a numeric value which is displayed inside the results pane of the SQL application. SQL also offers several built-in functions to perform what is known as aggregate data calculations against a table or a specific table column.

AVG() -- Returns the average value of a stated column. COUNT(*) -- Returns a count of the number of rows of table. SUM() -- Returns the sum of a given column.

Using one of the following functions also returns a numeric value:


SELECT COUNT(*) AS "Number of Orders"FROM orders;


SELECT SUM(quantity)AS "Total Number of Items Purchased"FROM orders;


SELECT AVG(quantity) AS "Average Number of Items Purchased"FROM orders;

We can also combine these queries into a single query so that the results are viewable all at once.

SQL Code:

USE mydatabase;

SELECT COUNT(*) AS "Number of Orders",SUM(quantity)AS "Total Number of Items Purchased",AVG(quantity)AS "Average Number of Items Purchased"FROM orders;

SQL - Date ExpressionsAs the name suggests, Date Expressions return date/time values.

GetDate() -- Returns the current date/time. Current_Timestamp -- Returns the current timestamp.

Date expressions as you may have guessed, return date values. We will be taking a closer look at date expressions later on in this tutorial. Stay tuned.


SELECT Current_Timestamp;SELECT GETDATE();

Documents

SQL Tutorial