Table of Contents
- Microsoft SQL Server Defined
- Microsoft SQL Server Features
- Microsoft SQL Server Administration
Microsoft SQL Server Programming
- An Outline for Development
- Database Services
- Database Objects: Databases
- Database Objects: Tables
- Database Objects: Table Relationships
- Database Objects: Keys
- Database Objects: Constraints
- Database Objects: Data Types
- Database Objects: Views
- Database Objects: Stored Procedures
- Database Objects: Indexes
- Database Objects: User Defined Functions
- Database Objects: Triggers
- Database Design: Requirements, Entities, and Attributes
- Business Process Model Notation (BPMN) and the Data Professional
- Business Questions for Database Design, Part One
- Business Questions for Database Design, Part Two
- Database Design: Finalizing Requirements and Defining Relationships
- Database Design: Creating an Entity Relationship Diagram
- Database Design: The Logical ERD
- Database Design: Adjusting The Model
- Database Design: Normalizing the Model
- Creating The Physical Model
- Database Design: Changing Attributes to Columns
- Database Design: Creating The Physical Database
- Database Design Example: Curriculum Vitae
- The SQL Server Sample Databases
- The SQL Server Sample Databases: pubs
- The SQL Server Sample Databases: NorthWind
- The SQL Server Sample Databases: AdventureWorks
- The SQL Server Sample Databases: Adventureworks Derivatives
- UniversalDB: The Demo and Testing Database, Part 1
- UniversalDB: The Demo and Testing Database, Part 2
- UniversalDB: The Demo and Testing Database, Part 3
- UniversalDB: The Demo and Testing Database, Part 4
- Getting Started with Transact-SQL
- Transact-SQL: Data Definition Language (DDL) Basics
- Transact-SQL: Limiting Results
- Transact-SQL: More Operators
- Transact-SQL: Ordering and Aggregating Data
- Transact-SQL: Subqueries
- Transact-SQL: Joins
- Transact-SQL: Complex Joins - Building a View with Multiple JOINs
- Transact-SQL: Inserts, Updates, and Deletes
- An Introduction to the CLR in SQL Server 2005
- Design Elements Part 1: Programming Flow Overview, Code Format and Commenting your Code
- Design Elements Part 2: Controlling SQL's Scope
- Design Elements Part 3: Error Handling
- Design Elements Part 4: Variables
- Design Elements Part 5: Where Does The Code Live?
- Design Elements Part 6: Math Operators and Functions
- Design Elements Part 7: Statistical Functions
- Design Elements Part 8: Summarization Statistical Algorithms
- Design Elements Part 9:Representing Data with Statistical Algorithms
- Design Elements Part 10: Interpreting the Data—Regression
- Design Elements Part 11: String Manipulation
- Design Elements Part 12: Loops
- Design Elements Part 13: Recursion
- Design Elements Part 14: Arrays
- Design Elements Part 15: Event-Driven Programming Vs. Scheduled Processes
- Design Elements Part 16: Event-Driven Programming
- Design Elements Part 17: Program Flow
- Forming Queries Part 1: Design
- Forming Queries Part 2: Query Basics
- Forming Queries Part 3: Query Optimization
- Forming Queries Part 4: SET Options
- Forming Queries Part 5: Table Optimization Hints
- Using SQL Server Templates
- Transact-SQL Unit Testing
- Index Tuning Wizard
- Unicode and SQL Server
- SQL Server Development Tools
- The SQL Server Transact-SQL Debugger
- The Transact-SQL Debugger, Part 2
- Basic Troubleshooting for Transact-SQL Code
- An Introduction to Spatial Data in SQL Server 2008
- Performance Tuning
- Practical Applications
- Professional Development
- Application Architecture Assessments
- Business Intelligence
- Tips and Troubleshooting
- Additional Resources
Design Elements Part 17: Program Flow
Last updated Mar 28, 2003.
Today I'll bring some of the Design Elements concepts together and cover program flow. Strictly speaking, program flow is the control over the direction the parts of the programs take based on loops, conditions, and other branching statements. I'm going to broaden the topic a bit to include some simple diagram techniques. Along the way, I'll introduce a couple of branching statements that I haven't covered yet.
Let's begin by creating a simple program requirement. First, we'll review the branching and block syntax we have available, then I'll show you some primitive diagramming for the program flow, and finally I'll implement that flow in a small section of code.
In the tutorial on An Outline for Development, I explained the phases for creating a program:
Understand the goal or problem as completely as possible
Comment the process
Code the comments
Optimize the code
As part of the first phase, it's important to get the system's requirements from the users, and communicating the design for the program to the developers. This is a necessary step, even if that developer is you. You can also use this design to show your understanding of the process to the users. >From the design you create, you can move on to the second phase of the outline above where you comment the code.
Some people work well from textual descriptions. Most of us, however, work better from a graphical representation. We see patterns that aren't easily explained with text alone. It's not difficult; basically, all you need are a few symbols that indicate the flow of the process and some text to explain the symbols.
Several tools have evolved over the years to diagram program flow. These systems range in complexity from three simple shapes (circle, square, and diamond) all the way to a full graphical vocabulary like you see in Universal Markup Language (UML). In this tutorial, I'll use fairly primitive symbols. I'll do that since the example I'll show doesn't require more than that, and I feel that the fewer the symbols the better. In other words, you should use the minimum number of symbols necessary to make the meaning clear.
Before we begin, let's review the syntax SQL provides for program flow, and how it affects programming in T-SQL. I've covered most of these constructs in other articles, and I promised to show you when they all come together. It is here in Program Flow where that starts to happen.
To begin, we have (appropriately enough) the BEGIN...END statements. Anything between these words is treated as a code block, and you can provide error trapping and other branching inside a block.
The WHILE statement creates a condition for a loop. WHILE causes the system to repeat the code block until the condition is met, or until a BREAK statement is encountered. If needed, the CONTINUE statement can restart the BREAK that is created. To leave a block no matter what a result is, you can use the RETURN statement.
Similar to the WHILE statement is the IF...ELSE statement. The IF statement tests a condition and then carries out the program statements that follow, or runs the code behind the ELSE statement.
To branch off to another section of code, you can use the GOTO statement, along with a Label. To create a label, you simply type a single word (or join two or more words with underlines) and end it with a colon, like this:
A new statement that I haven't covered yet is the CASE statement. This statement allows you to create one SELECT, and get different responses based on a test. Here's the syntax:
SELECT 'SomeTextHere' = CASE WHEN fieldname IS NULL THEN 'StringYouWant' WHEN fieldname = 'VariableOrValue' THEN 'OtherStringYouWant' WHEN fieldname < 'VariableOrValue' THEN 'OtherStringYouWant' WHEN fieldname > 'VariableOrValue' THEN 'OtherStringYouWant' WHEN fieldname <> 'VariableOrValue' THEN 'OtherStringYouWant' ELSE 'WhateverCatchesTheRest' END
You would replace the fieldname with a field you're testing against. You can set the return value of the CASE statement with a string (as I have here) or a constant, even a variable. Each one can be different. Just remember to close it all out with an END. We'll see this in our example code a little later.
Finally, there's the WAITFOR statement. I haven't covered this statement until now, so let's take a moment and see it in action.
The WAITFOR statement is clock-based. You can specify an absolute time like this:
USE master /* Backs up the pubs database at 10:00 in the morning */ BEGIN WAITFOR TIME '10:00' BACKUP DATABASE pubs TO DISK = 'c:\temp\pubs.bak' END
Or you can have the command run relatively to the current time, like this:
USE master /* Backs up the pubs database in one minute */ BEGIN WAITFOR DELAY '00:01:00' BACKUP DATABASE pubs TO DISK = 'c:\temp\pubs.bak' END
These constructs are the tools we have to work with so far, now let's explore the graphical tools that reference them.
As I mentioned, we'll use fairly simple graphical components to illustrate the program flow. You can use specialized tools that create flowcharts, or you can use something as simple as a graphics program to create the model, since it only contains three shapes and one arrow type. Many Microsoft-centric shops use Visio, but the shapes I'm showing here were created with Open Office's presentation package.
There are only three shapes I'll use: a circle, a box, and a diamond. I'll show the program direction with arrows.
In this simple schema, one circle represents the start and another represents the end of the process. A square represents a step, and the diamond represents a choice or branching operation.
In our fictional program, the requirements have been presented to us as follows:
"We need a stored procedure written that accepts an author's name and a book title as an input. The stored procedure will determine if the author wrote that book or another, or no books at all. If the author wrote the book passed as a variable, then the stored procedure should return the word 'Yes' as the answer, and if that author wrote another book that title should be returned. If the name provided isn't in the database, the words 'Not on File' should be returned."
I explained how to break this kind of request down into proper requirements in the article called Database Design: Requirements, Entities, and Attributes. In this exercise, we'll assume that's been done. I'll create a flow diagram of how the stored procedure might be written. You might wonder if I really take the time to flow-chart something this trivial (or whether it even needs a stored procedure at all!). The answer is... well, sometimes. Most of the time, you wouldn't need a full diagram for something as simple as this, but if this request is part of a larger design, you certainly would. For this exercise, a simple diagram is useful to illustrate the process of creating a graphical representation of the stored procedure.
Let's begin with a circle. Since it's obvious that the stored procedure has a start point, I use the start point to indicate any input parameters that are required.
Notice that I've also included the type of input that I'll receive from the calling program. You don't have to provide this information here, but it certainly can help in the flow diagram since a string of characters has more branching operations than a bit data type.
You can show as much granularity as you like. I normally include the SELECT statement operation, but you could go directly to the decision element.
Once I locate the author's name (if it exists) I have a decision to make. Based on the results of the query, there are three possible outcomes. One is that the author wrote that title, in which case I'll return the string "Yes". If the author wrote another book, I'll have to find out what the title is and then provide that title as the return string. Finally, if the author doesn't exist, I'll return the string "Not on File" to the calling program. Here's what I came up with:
Notice that the diamond contains a question. The answers are shown to the right, and the circles at the end of each decision contain the strings that are returned to the calling program based on the decision.
If the results of the decision create a situation wherein more steps are needed, a combination of boxes would illustrate those steps. Eventually, the steps culminate in a return or end value, or re-join another step or decision. Using arrows, the developer can trace what results are needed based on the results he or she gets from their queries.
What we're really doing here is abstracting the process so that both the developer and the business user can see the logic. This helps you avoid confusion and frustration on everyone's part.
Based on what I have here, here's one possible stored procedure:
USE pubs GO CREATE PROCEDURE FindAuthor @firstname varchar(18) , @lastname varchar(30) , @titlename varchar(100) AS SELECT 'TitleName' = CASE WHEN t.title IS NULL THEN 'Not on File' WHEN t.title = @titlename THEN 'Yes' ELSE title END FROM authors a INNER JOIN titleauthor ta ON a.au_id = ta.au_id INNER JOIN titles t ON t.title_id = ta.title_id WHERE au_fname LIKE @firstname AND au_lname LIKE @lastname
You can see the results with this statement:
EXEC FindAuthor 'Johnson', 'White', 'Prolonged Data Deprivation: Four Case Studies'
There's one fatal flaw in the original requirement. It seems that an author can write more than one book. How could we handle that? We'll leave that discussion to another time.
Want to learn more about UML? Paul Evitts has a great introduction to the elements it contains.
InformIT Tutorials and Sample Chapters
The seminal work on design patterns is still the best: http://www.informit.com/title/0201633612