Donald D. Chamberlin
Raymond F. Boyce
|Stable release||SQL:2008 (2008)|
|Typing discipline||Static, strong|
|Dialects||SQL-86, SQL-89, SQL-92, SQL:1999, SQL:2003, SQL:2008|
|Influenced||Agena, CQL, LINQ, Windows PowerShell|
|Usual file extensions||.sql|
|Internet media type||
|Initial release||1986 (1986)|
|Latest release||SQL:2008 / 2008; 2 years ago (2008)|
|Type of format||Database|
SQL (officially pronounced /??skju?'?l/ like "S-Q-L" but often pronounced /'si?kw?l/ like "sequel"), often referred to as Structured Query Language, is a database computer language designed for managing data in relational database management systems (RDBMS), and originally based upon relational algebra and calculus. Its scope includes data insert, query, update and delete, schema creation and modification, and data access control. SQL was one of the first commercial languages for Edgar F. Codd's relational model, as described in his influential 1970 paper, "A Relational Model of Data for Large Shared Data Banks". Despite not adhering to the relational model as described by Codd, it became the most widely used database language.
SQL was developed at IBM by Donald D. Chamberlin and Raymond F. Boyce in the early 1970s. This version, initially called SEQUEL (Structured English Query Language), was designed to manipulate and retrieve data stored in IBM's original quasi-relational database management system, System R, which a group at IBM San Jose Research Laboratory had developed during the 1970s. The acronym SEQUEL was later changed to SQL because "SEQUEL" was a trademark of the UK-based Hawker Siddeley aircraft company.
The first Relational Database Management System (RDBMS) was RDMS, developed at MIT in the early 1970s, soon followed by Ingres, developed in 1974 at U.C. Berkeley. Ingres implemented a query language known as QUEL, which was later supplanted in the marketplace by SQL.
In the late 1970s, Relational Software, Inc. (now Oracle Corporation) saw the potential of the concepts described by Codd, Chamberlin, and Boyce and developed their own SQL-based RDBMS with aspirations of selling it to the U.S. Navy, Central Intelligence Agency, and other U.S. government agencies. In June 1979, Relational Software, Inc. introduced the first commercially available implementation of SQL, Oracle V2 (Version2) for VAX computers. Oracle V2 beat IBM's August release of the System/38 RDBMS to market by a few weeks.
After testing SQL at customer test sites to determine the usefulness and practicality of the system, IBM began developing commercial products based on their System R prototype including System/38, SQL/DS, and DB2, which were commercially available in 1979, 1981, and 1983, respectively.
The SQL language is sub-divided into several language elements, including:
The most common operation in SQL is the query, which is performed with the declarative
SELECT retrieves data from one or more tables, or expressions. Standard
SELECT statements have no persistent effects on the database. Some non-standard implementations of
SELECT can have persistent effects, such as the
SELECT INTO syntax that exists in some databases.
Queries allow the user to describe desired data, leaving the database management system (DBMS) responsible for planning, optimizing, and performing the physical operations necessary to produce that result as it chooses.
A query includes a list of columns to be included in the final result immediately following the
SELECT keyword. An asterisk ("
*") can also be used to specify that the query should return all columns of the queried tables.
SELECT is the most complex statement in SQL, with optional keywords and clauses that include:
FROMclause which indicates the table(s) from which data is to be retrieved. The
FROMclause can include optional
JOINsubclauses to specify the rules for joining tables.
WHEREclause includes a comparison predicate, which restricts the rows returned by the query. The
WHEREclause eliminates all rows from the result set for which the comparison predicate does not evaluate to True.
GROUP BYclause is used to project rows having common values into a smaller set of rows.
GROUP BYis often used in conjunction with SQL aggregation functions or to eliminate duplicate rows from a result set. The
WHEREclause is applied before the
HAVINGclause includes a predicate used to filter rows resulting from the
GROUP BYclause. Because it acts on the results of the
GROUP BYclause, aggregation functions can be used in the
ORDER BYclause identifies which columns are used to sort the resulting data, and in which direction they should be sorted (options are ascending or descending). Without an
ORDER BYclause, the order of rows returned by an SQL query is undefined.
The following is an example of a
SELECT query that returns a list of expensive books. The query retrieves all rows from the Book table in which the price column contains a value greater than 100.00. The result is sorted in ascending order by title. The asterisk (*) in the select list indicates that all columns of the Book table should be included in the result set.
SELECT * FROM Book WHERE price > 100.00 ORDER BY title;
The example below demonstrates a query of multiple tables, grouping, and aggregation, by returning a list of books and the number of authors associated with each book.
SELECT Book.title, count(*) AS Authors FROM Book JOIN Book_author ON Book.isbn = Book_author.isbn GROUP BY Book.title;
Example output might resemble the following:
Title Authors ---------------------- ------- SQL Examples and Guide 4 The Joy of SQL 1 An Introduction to SQL 2 Pitfalls of SQL 1
Under the precondition that isbn is the only common column name of the two tables and that a column named title only exists in the Books table, the above query could be rewritten in the following form:
SELECT title, count(*) AS Authors FROM Book NATURAL JOIN Book_author GROUP BY title;
However, many vendors either do not support this approach, or require certain column naming conventions in order for natural joins to work effectively.
SQL includes operators and functions for calculating values on stored values. SQL allows the use of expressions in the select list to project data, as in the following example which returns a list of books that cost more than 100.00 with an additional sales_tax column containing a sales tax figure calculated at 6% of the price.
SELECT isbn, title, price, price * 0.06 AS sales_tax FROM Book WHERE price > 100.00 ORDER BY title;
The idea of Null was introduced into SQL to handle missing information in the relational model. The introduction of Null (or Unknown) along with True and False is the foundation of three-valued logic. Null does not have a value (and is not a member of any data domain) but is rather a placeholder or "mark" for missing information. Therefore comparisons with Null can never result in either True or False but always in the third logical result.
SQL uses Null to handle missing information. It supports three-valued logic (3VL) and the rules governing SQL three-valued logic are shown below (p and q represent logical states). The word NULL is also a reserved keyword in SQL, used to identify the Null special marker.
Additionally, since SQL operators return Unknown when comparing anything with Null, SQL provides two Null-specific comparison predicates:
IS NULL and
IS NOT NULL test whether data is or is not Null.
Note that SQL returns only results for which the WHERE clause returns a value of True; i.e. it excludes results with values of False, but also those whose value is Unknown.
Universal quantification is not explicitly supported by SQL, and must be worked out as a negated existential quantification.
There is also the "<row value expression> IS DISTINCT FROM <row value expression>" infixed comparison operator which returns TRUE unless both operands are equal or both are NULL. Likewise, IS NOT DISTINCT FROM is defined as "NOT (<row value expression> IS DISTINCT FROM <row value expression>)".
The Data Manipulation Language (DML) is the subset of SQL used to add, update and delete data:
INSERTadds rows (formally tuples) to an existing table, e.g.,:
INSERT INTO My_table (field1, field2, field3) VALUES ('test', 'N', NULL);
UPDATEmodifies a set of existing table rows, e.g.,:
UPDATE My_table SET field1 = 'updated value' WHERE field2 = 'N';
DELETEremoves existing rows from a table, e.g.,:
DELETE FROM My_table WHERE field2 = 'N';
MERGEis used to combine the data of multiple tables. It combines the
UPDATEelements. It is defined in the SQL:2003 standard; prior to that, some databases provided similar functionality via different syntax, sometimes called "upsert".
Transactions, if available, wrap DML operations:
BEGIN WORK, or
BEGIN TRANSACTION, depending on SQL dialect) mark the start of a database transaction, which either completes entirely or not at all.
SAVEPOINT) save the state of the database at the current point in transaction
CREATE TABLE tbl_1(id int); INSERT INTO tbl_1(id) value(1); INSERT INTO tbl_1(id) value(2); COMMIT; UPDATE tbl_1 SET id=200 WHERE id=1; SAVEPOINT id_1upd; UPDATE tbl_1 SET id=1000 WHERE id=2; ROLLBACK TO id_1upd; SELECT id FROM tbl_1;
COMMITcauses all data changes in a transaction to be made permanent.
ROLLBACKcauses all data changes since the last
ROLLBACKto be discarded, leaving the state of the data as it was prior to those changes.
COMMIT statement completes, the transaction's changes cannot be rolled back.
ROLLBACK terminate the current transaction and release data locks. In the absence of a
START TRANSACTION or similar statement, the semantics of SQL are implementation-dependent. Example: A classic bank transfer of funds transaction.
START TRANSACTION; UPDATE Account SET amount=amount-200 WHERE account_number=1234; UPDATE Account SET amount=amount+200 WHERE account_number=2345; IF ERRORS=0 COMMIT; IF ERRORS<>0 ROLLBACK;
The Data Definition Language (DDL) manages table and index structure. The most basic items of DDL are the
CREATEcreates an object (a table, for example) in the database, e.g.,:
CREATE TABLE My_table( my_field1 INT, my_field2 VARCHAR(50), my_field3 DATE NOT NULL, PRIMARY KEY (my_field1, my_field2) );
ALTERmodifies the structure of an existing object in various ways, for example, adding a column to an existing table or a constraint, e.g.,:
ALTER TABLE My_table ADD my_field4 NUMBER(3) NOT NULL;
TRUNCATEdeletes all data from a table in a very fast way, deleting the data inside the table and not the table itself. It usually implies a subsequent COMMIT operation, i.e., it cannot be rolled back.
TRUNCATE TABLE My_table;
DROPdeletes an object in the database, usually irretrievably, i.e., it cannot be rolled back, e.g.,:
DROP TABLE My_table;
Each column in an SQL table declares the type(s) that column may contain. ANSI SQL includes the following datatypes.
CHAR(n)- fixed-width n-character string, padded with spaces as needed
VARCHAR(n)- variable-width string with a maximum size of n characters
NCHAR(n)- fixed width string supporting an international character set
NATIONAL CHARACTER VARYING(n)or
BIT(n)- an array of n bits
BIT VARYING(n)- an array of up to n bits
SQL provides a function to round numerics or dates, called
TRUNC (in Informix, DB2, PostgreSQL, Oracle and MySQL) or
ROUND (in Informix, Sybase, Oracle, PostgreSQL and Microsoft SQL Server)
DATE- for date values (e.g.,
TIME- for time values (e.g.,
14:55:37). The granularity of the time value is usually a tick (100 nanoseconds).
TIME WITH TIME ZONEor
TIMESTAMP- the same as
TIME, but including details about the time zone in question.
TIMESTAMP- This is a
TIMEput together in one variable (e.g.,
TIMESTAMP WITH TIME ZONEor
TIMESTAMPTZ- the same as
TIMESTAMP, but including details about the time zone in question.
SQL provides several functions for generating a date / time variable out of a date / time string (
TO_TIMESTAMP), as well as for extracting the respective members (seconds, for instance) of such variables. The current system date / time of the database server can be called by using functions like
The Data Control Language (DCL) authorizes users and groups of users to access and manipulate data. Its two main statements are:
GRANTauthorizes one or more users to perform an operation or a set of operations on an object.
REVOKEeliminates a grant, which may be the default grant.
GRANT SELECT, UPDATE ON My_table TO some_user, another_user; REVOKE SELECT, UPDATE ON My_table FROM some_user, another_user;
SQL is designed for a specific purpose: to query data contained in a relational database. SQL is a set-based, declarative query language, not an imperative language such as C or BASIC. However, there are extensions to Standard SQL which add procedural programming language functionality, such as control-of-flow constructs. These include:
|ANSI/ISO Standard||SQL/PSM||SQL/Persistent Stored Modules|
|IBM||SQL PL||SQL Procedural Language (implements SQL/PSM)|
|Mimer SQL||SQL/PSM||SQL/Persistent Stored Module (implements SQL/PSM)|
|MySQL||SQL/PSM||SQL/Persistent Stored Module (implements SQL/PSM)|
|Oracle||PL/SQL||Procedural Language/SQL (based on Ada)|
|PostgreSQL||PL/pgSQL||Procedural Language/PostgreSQL Structured Query Language (based on Oracle PL/SQL)|
|PostgreSQL||PL/PSM||Procedural Language/Persistent Stored Modules (implements SQL/PSM)|
In addition to the standard SQL/PSM extensions and proprietary SQL extensions, procedural and object-oriented programmability is available on many SQL platforms via DBMS integration with other languages. The SQL standard defines SQL/JRT extensions (SQL Routines and Types for the Java Programming Language) to support Java code in SQL databases. SQL Server 2005 uses the SQLCLR (SQL Server Common Language Runtime) to host managed .NET assemblies in the database, while prior versions of SQL Server were restricted to using unmanaged extended stored procedures which were primarily written in C. Other database platforms, like MySQL and Postgres, allow functions to be written in a wide variety of languages including Perl, Python, Tcl, and C.
SQL is a declarative computer language intended for use with relational databases. Many of the original SQL features were inspired by, but violated, the semantics of the relational model and its tuple calculus realization. Recent extensions to SQL achieved relational completeness, but have worsened the violations, as documented in The Third Manifesto. Therefore, it cannot be considered relational in any significant sense, but is still widely called relational due to differentiation to other, pre-relational database languages which never intended to implement the relational model; due to its historical origin; and due to the use of the relational term as propaganda by product vendors.
Other criticisms of SQL include:
NULLs, and comparison case sensitivity vary from vendor to vendor. A particular exception is PostgreSQL, which strives for compliance, and SQLite, which strives to follow PostgreSQL.
WHEREclauses are mistyped. Cartesian joins are so rarely used in practice that requiring an explicit
CARTESIANkeyword may be warranted. (SQL 1992 introduced the
CROSS JOINkeyword that allows the user to make clear that a Cartesian join is intended, but the shorthand "comma-join" with no predicate is still acceptable syntax, which still invites the same mistake.)
WHEREon an update or delete, thereby affecting more rows in a table than desired. (A work-around is to use transactions or habitually type in the WHERE clause first, then fill in the rest later.)
Popular implementations of SQL commonly omit support for basic features of Standard SQL, such as the
TIME data types. The most obvious such examples, and incidentally the most popular commercial, proprietary SQL DBMSs, are Oracle (whose
DATE behaves as
DATETIME, and lacks a
TIME type) and the MS SQL Server. As a result, SQL code can rarely be ported between database systems without modifications.
There are several reasons for this lack of portability between database systems:
SQL was adopted as a standard by the American National Standards Institute (ANSI) in 1986 as SQL-86 and International Organization for Standardization (ISO) in 1987. The original SQL standard declared that the official pronunciation for SQL is "es queue el". Many English-speaking database professionals still use the nonstandard pronunciation /'si?kw?l/ (like the word "sequel").
Until 1996, the National Institute of Standards and Technology (NIST) data management standards program certified SQL DBMS compliance with the SQL standard. Vendors now self-certify the compliance of their products.
The SQL standard has gone through a number of revisions, as shown below:
|1986||SQL-86||SQL-87||First formalized by ANSI.|
|1989||SQL-89||FIPS 127-1||Minor revision, adopted as FIPS 127-1.|
|1992||SQL-92||SQL2, FIPS 127-2||Major revision (ISO 9075), Entry Level SQL-92 adopted as FIPS 127-2.|
|1999||SQL:1999||SQL3||Added regular expression matching, recursive queries, triggers, support for procedural and control-of-flow statements, non-scalar types, and some object-oriented features.|
|2003||SQL:2003||SQL 2003||Introduced XML-related features, window functions, standardized sequences, and columns with auto-generated values (including identity-columns).|
|2006||SQL:2006||SQL 2006||ISO/IEC 9075-14:2006 defines ways in which SQL can be used in conjunction with XML. It defines ways of importing and storing XML data in an SQL database, manipulating it within the database and publishing both XML and conventional SQL-data in XML form. In addition, it enables applications to integrate into their SQL code the use of XQuery, the XML Query Language published by the World Wide Web Consortium (W3C), to concurrently access ordinary SQL-data and XML documents.|
|2008||SQL:2008||SQL 2008||Legalizes ORDER BY outside cursor definitions. Adds INSTEAD OF triggers. Adds the TRUNCATE statement.|
Interested parties may purchase SQL standards documents from ISO or ANSI. A draft of SQL:2008 is freely available as a zip archive.
The SQL standard is divided into several parts, including:
SQL Framework, provides logical concept
SQL/Foundation, defined in ISO/IEC 9075, Part 2. This part of the standard contains the most central elements of the language. It consists of both mandatory and optional features.
The SQL/Bindings, specifies how SQL is to be bound to variable host languages, excluding Java.
The SQL/CLI, or Call-Level Interface, part is defined in ISO/IEC 9075, Part 3. SQL/CLI defines common interfacing components (structures and procedures) that can be used to execute SQL statements from applications written in other programming languages. SQL/CLI is defined in such a way that SQL statements and SQL/CLI procedure calls are treated as separate from the calling application's source code. Open Database Connectivity is a well-known superset of SQL/CLI. This part of the standard consists solely of mandatory features.
The SQL/PSM, or Persistent Stored Modules, part is defined by ISO/IEC 9075, Part 4. SQL/PSM standardizes procedural extensions for SQL, including flow of control, condition handling, statement condition signals and resignals, cursors and local variables, and assignment of expressions to variables and parameters. In addition, SQL/PSM formalizes declaration and maintenance of persistent database language routines (e.g., "stored procedures"). This part of the standard consists solely of optional features.
The SQL/MED, or Management of External Data, part is defined by ISO/IEC 9075, Part 9. SQL/MED provides extensions to SQL that define foreign-data wrappers and datalink types to allow SQL to manage external data. External data is data that is accessible to, but not managed by, an SQL-based DBMS. This part of the standard consists solely of optional features.
The SQL/OLB, or Object Language Bindings, part is defined by ISO/IEC 9075, Part 10. SQL/OLB defines the syntax and symantics of SQLJ, which is SQL embedded in Java. The standard also describes mechanisms to ensure binary portability of SQLJ applications, and specifies various Java packages and their contained classes. This part of the standard consists solely of optional features.
The SQL/MM (Multimedia), This extends SQL to deal intelligently with large, complex and sometimes streaming items of data, such as video, audio and spatial data.
The SQL/Schemata, or Information and Definition Schemas, part is defined by ISO/IEC 9075, Part 11. SQL/Schemata defines the Information Schema and Definition Schema, providing a common set of tools to make SQL databases and objects self-describing. These tools include the SQL object identifier, structure and integrity constraints, security and authorization specifications, features and packages of ISO/IEC 9075, support of features provided by SQL-based DBMS implementations, SQL-based DBMS implementation information and sizing items, and the values supported by the DBMS implementations. This part of the standard contains both mandatory and optional features.
The SQL/JRT, or SQL Routines and Types for the Java Programming Language, part is defined by ISO/IEC 9075, Part 13. SQL/JRT specifies the ability to invoke static Java methods as routines from within SQL applications. It also calls for the ability to use Java classes as SQL structured user-defined types. This part of the standard consists solely of optional features.
The SQL/XML, or XML-Related Specifications, part is defined by ISO/IEC 9075, Part 14. SQL/XML specifies SQL-based extensions for using XML in conjunction with SQL. The XML data type is introduced, as well as several routines, functions, and XML-to-SQL data type mappings to support manipulation and storage of XML in an SQL database. This part of the standard consists solely of optional features.