Cheat Sheet

SQL All-in-One For Dummies

SQL is a popular and useful programming language. You can make SQL even more useful if you know the phases of SQL development, the criteria for normal forms, the data types used by SQL, a little bit about set and value functions, as well as some tips on how to filter tables with WHERE clauses.

Phases of SQL System Development

In developing any system, you start at the beginning and go through to the end, and it's no different with SQL. The following list shows you what to consider at each phase of the SQL development lifecycle:

  • Definition Phase: Precisely define the problem to be solved, its magnitude, and who will work on it.

  • Requirements Phase: Develop a detailed description of exactly what the development effort will produce. Gather all relevant information and put it into a requirements document (Statement of Requirements). Get client signoff.

  • Evaluation Phase: Determine exactly how you will meet the requirements. What tools will you use? How will you deploy your development team? Determine whether the job is doable within time and budget constraints.

  • Design Phase: Create a database model and then design a database and database application that satisfy the terms of the requirements document.

  • Implementation Phase: Build the database and the database application. Include copious documentation within the code and in external documents.

  • Final Documentation and Testing Phase: Give the database and application a tough workout. Hit the system with every conceivable input condition and a few inconceivable ones. Try to overload it. See where it breaks. When it breaks, send it back to the implementers or even back to the designers. Document everything.

  • Maintenance Phase: Fix latent bugs as they arise. Provide updates and enhancements called for by the client.

SQL Criteria for Normal Forms

In SQL, normal forms are defining characteristics of relational databases. SQL forms get classified according to the types of modification anomalies they're subject to. First, second, and third normal forms (1NF, 2NF, 3NF) serve as remedies to the three main sources of modification anomalies.

The normal forms are nested in the sense that a table that's in 2NF is automatically also in 1NF. Similarly, a table in 3NF is automatically in 2NF, and so on. For most practical applications, putting a database in 3NF is sufficient to ensure a high degree of integrity. To be absolutely sure of its integrity, you must put the database into DK/NF.

The following lists lay out the criteria for each form:

First Normal Form (1NF):

  • Table must be two-dimensional, with rows and columns.

  • Each row contains data that pertains to one thing or one portion of a thing.

  • Each column contains data for a single attribute of the thing being described.

  • Each cell (intersection of row and column) of the table must be single-valued.

  • All entries in a column must be of the same kind.

  • Each column must have a unique name.

  • No two rows may be identical.

  • The order of the columns and of the rows does not matter.

Second Normal Form (2NF):

  • Table must be in first normal form (1NF).

  • All nonkey attributes (columns) must be dependent on the entire key.

Third Normal Form (3NF):

  • Table must be in second normal form (2NF).

  • Table has no transitive dependencies.

Domain-Key Normal Form (DK/NF):

  • Every constraint on the table is a logical consequence of the definition of keys and domains.

SQL Data Types

Depending on their histories, different SQL implementations support a variety of data types. The SQL specification recognizes nine predefined general types, shown in the lists below

Exact Numerics:

  • INTEGER

  • SMALLINT

  • BIGINT

  • NUMERIC

  • DECIMAL

Approximate Numerics:

  • REAL

  • DOUBLE PRECISION

  • FLOAT

Boolean:

  • BOOLEAN

Character Strings:

  • CHARACTER

  • CHARACTER VARYING (VARCHAR)

  • NATIONAL CHARACTER

  • NATIONAL CHARACTER VARYING

Datetimes:

  • DATE

  • TIME

  • TIMESTAMP

  • TIME WITH TIMEZONE

  • TIMESTAMP WITH TIMEZONE

Intervals:

  • INTERVAL DAY

  • INTERVAL YEAR

Large Objects:

  • BLOB

  • CLOB

Collection Types:

  • ARRAY

  • MULTISET

Other Types:

  • ROW

  • XML

SQL Value Functions

You use SQL value expressions to combine two or more values. Several kinds of SQL value expressions exist, corresponding to the different data types; the following tables list string, numeric, and datetime values, functions, and effects:

String Value Functions
Function Effect
SUBSTRING Extracts a substring from a source string
UPPER Converts a character string to all uppercase
LOWER Converts a character string to all lowercase
TRIM Trims off leading or trailing blanks
TRANSLATE Transforms a source string from one character set to another
CONVERT Same as TRANSLATE. (It transforms a source string from one character set to another.)
Numeric Value Functions
Function Effect
POSITION Returns the starting position of a target string within a source string
CHARACTER_LENGTH Returns the number of characters in a string
OCTET_LENGTH Returns the number of octets (bytes) in a character string
EXTRACT Extracts a single field from a datetime or interval
Datetime Value Functions
Function Effect
CURRENT_DATE Returns the current date
CURRENT_TIME(p) Returns the current time; (p) is precision of seconds
CURRENT_TIMESTAMP(p) Returns the current date and the current time; (p) is precision of seconds

SQL Set Functions

Sometimes, the information that you want to extract from an SQL table doesn't relate to individual rows but rather to sets of rows. SQL provides five set (or aggregate) functions to deal with such situations — COUNT, MAX, MIN, SUM, and AVG. Each function performs an action that draws data from a set of rows rather than from a single row:

COUNT Returns the number of rows in the specified table
MAX Returns the maximum value that occurs in the specified table
MIN Returns the minimum value that occurs in the specified table
SUM Adds up the values in a specified column
AVG Returns the average of all the values in the specified column

SQL WHERE Clause Predicates

The WHERE clause is an SQL filter that passes rows that meet the search condition and rejects rows that don't meet the condition. By including WHERE clauses in your SQL SELECT statements, you can restrict the rows that you place into the resulting table to those that satisfy specific conditions. The following table lists the predicates to use to filter for the information you want:

Comparison Predicates    
= Equal
<> Not equal
< Less than
<= Less than or equal
> Greater than
>= Greater than or equal
Other Predicates    
ALL BETWEEN
DISTINCT EXISTS
IN LIKE
MATCH NOT IN
NOT LIKE NULL
OVERLAPS SIMILAR
SOME, ANY UNIQUE
  • Add a Comment
  • Print
  • Share
blog comments powered by Disqus
Advertisement

Inside Dummies.com

Dummies.com Sweepstakes

Win $500. Easy.