View source for SQL syntax

{{Short description|Set of rules defining correctly structured programs}}

The '''[[Syntax (programming languages)|syntax]] of the [[SQL]] programming language''' is defined and maintained by [[ISO/IEC JTC 1/SC 32|ISO/IEC SC 32]] as part of ''ISO/IEC 9075''. This standard is not freely available. Despite the existence of the standard, SQL code is not completely portable among different [[Database|database systems]] without adjustments.

== Language elements ==

{{Image frame|content=
<math>
\left. \begin{array}{rl} \scriptstyle\mathtt{UPDATE~clause} & \{ \mathtt{UPDATE\ country}  \\ 
\scriptstyle\mathtt{SET~clause} & \{ \mathtt{SET\ population =~} \overbrace\mathtt{population + 1}^\mathtt{expression}  \\ 
\scriptstyle\mathtt{WHERE~clause} & \{ \mathtt{WHERE\ \underbrace{{name =} \overbrace{'USA'}^{expression}}_{predicate};} 
\end{array}
\right\}{\scriptstyle\texttt{statement}}
</math>
|width=500|caption=A chart showing several of the SQL language elements that compose a single statement. This adds one to the population of the USA in the country table.}}

The SQL language is subdivided into several language elements, including:

* ''Keywords'' are words that are defined in the SQL language. They are either reserved (e.g. {{code|SELECT}}, {{code|COUNT}} and {{code|YEAR}}), or non-reserved (e.g. {{code|ASC}}, {{code|DOMAIN}} and {{code|KEY}}). List of [[SQL reserved words]].
* ''Identifiers'' are names on database objects, like [[Table (database)|tables]], columns and schemas. An identifier may not be equal to a reserved keyword, unless it is a delimited identifier. Delimited identifiers means identifiers enclosed in double quotation marks. They can contain characters normally not supported in SQL identifiers, and they can be identical to a reserved word, e.g. a column named {{code|YEAR}} is specified as {{code|"YEAR"}}.
** In [[MySQL]], double quotes are [[string literal]] delimiters by default instead. Enabling the {{code|ansi_quotes}} SQL mode enforces the SQL standard behavior. These can also be used regardless of this mode through backticks: {{code|`YEAR`}}.
* ''Clauses'', which are constituent components of [[Statement (computer science)|statements]] and queries. (In some cases, these are optional.)<ref name="ANSI/ISO/IEC">ANSI/ISO/IEC International Standard (IS). Database Language SQL—Part 2: Foundation (SQL/Foundation). 1999.</ref>
* ''Expressions'', which can produce either [[scalar (computing)|scalar]] values, or [[table (database)|tables]] consisting of [[column (database)|columns]] and [[row (database)|rows]] of data
* ''Predicates'', which specify conditions that can be evaluated to SQL [[ternary logic|three-valued logic (3VL)]] (true/false/unknown) or [[Boolean logic|Boolean]] [[truth value]]s and are used to limit the effects of statements and queries, or to change program flow.
* ''Queries'', which retrieve the data based on specific criteria. This is an important element of ''SQL''.
* ''Statements'', which may have a persistent effect on schemata and data, or may control [[Database transaction|transactions]], [[Control flow|program flow]], connections, [[Session (computer science)|sessions]], or diagnostics.
** SQL statements also include the [[semicolon]] (";") statement terminator. Though not required on every platform, it is defined as a standard part of the SQL grammar.
* ''[[Whitespace (computer science)|Insignificant whitespace]]'' is generally ignored in SQL statements and queries, making it easier to format SQL code for readability.

== Operators ==

{| class="wikitable"
|-
! Operator
! Description
! Example
|-
| <code>=</code>
| Equal to
| {{code|2=sql|1= Author = 'Alcott'}}
|-
| <code><></code>
| Not equal to (many dialects also accept <code>!=</code>)
| {{code|2=sql|1=Dept <> 'Sales'}}
|-
| <code>></code>
| Greater than
| {{code|2=sql|1=Hire_Date > '2012-01-31'}}
|-
| <code><</code>
| Less than
| {{code|2=sql|1=Bonus < 50000.00}}
|-
| <code>>=</code>
| Greater than or equal
| {{code|2=sql|1=Dependents >= 2}}
|-
| <code><=</code>
| Less than or equal
| {{code|2=sql|1=Rate <= 0.05}}
|-
| [[Where (SQL)#BETWEEN|{{code|2=sql|1=[NOT] BETWEEN [SYMMETRIC]}}]]
| Between an inclusive range. SYMMETRIC inverts the range bounds if the first is higher than the second.
| {{code|2=sql|1=Cost BETWEEN 100.00 AND 500.00}}
|-
|rowspan=2| [[Like (SQL)|{{code|2=sql|1=[NOT] LIKE [ESCAPE]}}]]
| Begins with a character pattern
| {{code|2=sql|1=Full_Name LIKE 'Will%'}}
|-
| Contains a character pattern
| {{code|2=sql|1=Full_Name LIKE '%Will%'}}
|-
| [[Where (SQL)#IN|{{code|2=sql|1=[NOT] IN}}]]
| Equal to one of multiple possible values
| {{code|2=sql|1=DeptCode IN (101, 103, 209)}}
|-
| [[SQL syntax#Null or three-valued logic (3VL)|{{code|2=sql|1=IS [NOT] NULL}}]]
| Compare to null (missing data)
| {{code|2=sql|1=Address IS NOT NULL}}
|-
| {{code|2=sql|1=IS [NOT] TRUE}} ''or'' {{code|2=sql|1=IS [NOT] FALSE}}
| Boolean truth value test
| {{code|2=sql|1=PaidVacation IS TRUE}}
|-
| [[SQL syntax#Null or three-valued logic (3VL)|{{code|2=sql|1=IS NOT DISTINCT FROM}}]]
| Is equal to value or both are nulls (missing data)
| {{code|2=sql|1=Debt IS NOT DISTINCT FROM - Receivables}}
|-
| {{code|2=sql|1=AS}}
| Used to change a column name when viewing results
| {{code|2=sql|1=SELECT employee AS department1}}
|}

Other operators have at times been suggested or implemented, such as the [[skyline operator]] (for finding only those rows that are not 'worse' than any others).

{{anchor|Conditional (CASE) expressions}}
SQL has the {{code|case}} expression, which was introduced in [[SQL-92]]. In its most general form, which is called a "searched case" in the SQL standard:

<syntaxhighlight lang="sql">
CASE WHEN n > 0
          THEN 'positive'
     WHEN n < 0
          THEN 'negative'
     ELSE 'zero'
END
</syntaxhighlight>

SQL tests {{code|WHEN}} conditions in the order they appear in the source. If the source does not specify an {{code|ELSE}} expression, SQL defaults to {{code|ELSE NULL}}. An abbreviated syntax called "simple case" can also be used:

<syntaxhighlight lang="sql">
CASE n WHEN 1
            THEN 'One'
       WHEN 2
            THEN 'Two'
       ELSE 'I cannot count that high'
END
</syntaxhighlight>

This syntax uses implicit equality comparisons, with [[SQL CASE|the usual caveats for comparing with NULL]].

There are two short forms for special {{code|CASE}} expressions: {{code|COALESCE}} and {{code|NULLIF}}.

The {{code|COALESCE}} expression returns the value of the first non-NULL operand, found by working from left to right, or NULL if all the operands equal NULL.

<syntaxhighlight lang="sql">
COALESCE(x1,x2)
</syntaxhighlight> 
is equivalent to:

<syntaxhighlight lang="sql">
CASE WHEN x1 IS NOT NULL THEN x1
     ELSE x2
END
</syntaxhighlight>

The {{code|NULLIF}} expression has two operands and returns NULL if the operands have the same value, otherwise it has the value of the first operand.

<syntaxhighlight lang="sql">
NULLIF(x1, x2)
</syntaxhighlight>
 
is equivalent to
<syntaxhighlight lang="sql">
CASE WHEN x1 = x2 THEN NULL ELSE x1 END
</syntaxhighlight>

== Comments ==
Standard SQL allows two formats for [[Comment (computer programming)|comments]]: {{code|-- comment}}, which is ended by the first [[newline]], and {{code|/* comment */}}, which can span multiple lines.

== Queries ==

The most common operation in SQL, the query, makes use of the declarative <code>[[Select (SQL)|SELECT]]</code> statement. {{code|SELECT}} retrieves data from one or more [[Table (database)|table]]s, or expressions. Standard {{code|SELECT}} statements have no persistent effects on the database. Some non-standard implementations of {{code|SELECT}} can have persistent effects, such as the {{code|SELECT INTO}} syntax provided in some databases.<ref name="ms-sql-select-into">{{cite web|url=https://learn.microsoft.com/en-us/sql/t-sql/queries/select-into-clause-transact-sql|title=SELECT - INTO Clause (Transact-SQL)|website=learn.microsoft.com|date=2023-05-23|access-date=2024-11-07}}</ref>

Queries allow the user to describe desired data, leaving the [[Database management system|database management system (DBMS)]] to carry out [[query plan|planning]], [[query optimizer|optimizing]], and performing the physical operations necessary to produce that result as it chooses.

A query includes a list of columns to include in the final result, normally immediately following the {{code|SELECT}} keyword. An asterisk ("<code>*</code>") can be used to specify that the query should return all columns of the queried tables. {{code|SELECT}} is the most complex statement in SQL, with optional keywords and clauses that include:

* The <code>[[From (SQL)|FROM]]</code> clause, which indicates the table(s) to retrieve data from. The {{code|FROM}} clause can include optional <code>[[Join (SQL)|JOIN]]</code> subclauses to specify the rules for joining tables.
* The <code>WHERE</code> clause includes a comparison predicate, which restricts the rows returned by the query. The {{code|WHERE}} clause eliminates all rows from the [[result set]] where the comparison predicate does not evaluate to True.
* The <code>[[Group by (SQL)|GROUP BY]]</code> clause projects rows having common values into a smaller set of rows.{{Clarify|reason=|date=September 2020}} {{code|GROUP BY}} is often used in conjunction with SQL aggregation functions or to eliminate duplicate rows from a result set. The {{code|WHERE}} clause is applied before the {{code|GROUP BY}} clause.
* The <code>[[Having (SQL)|HAVING]]</code> clause includes a predicate used to filter rows resulting from the {{code|GROUP BY}} clause. Because it acts on the results of the {{code|GROUP BY}} clause, aggregation functions can be used in the {{code|HAVING}} clause predicate.
* The <code>[[Order by (SQL)|ORDER BY]]</code> clause identifies which column[s] to use to sort the resulting data, and in which direction to sort them (ascending or descending). Without an {{code|ORDER BY}} clause, the order of rows returned by an SQL query is undefined.
* The <code>DISTINCT</code> keyword<ref>
{{cite book
| title                 = SAS 9.4 SQL Procedure User's Guide
| url                   = https://books.google.com/books?id=ESjMAAAAQBAJ
| publisher             = SAS Institute
| date      = 2013
| page                  = 248
| isbn                  = 9781612905686
| access-date            = 2015-10-21
| quote                 = Although the UNIQUE argument is identical to DISTINCT, it is not an ANSI standard.
}}
</ref> eliminates duplicate data.<ref>
{{cite book
| last1                 = Leon
| first1                = Alexis
| author-link1          = Alexis Leon
| last2                 = Leon
| first2                = Mathews
| year              = 1999
| chapter               = Eliminating duplicates - SELECT using DISTINCT
| title                 = SQL: A Complete Reference
| url                   = https://books.google.com/books?id=dmiPz2MMpfwC
| location              = New Delhi
| publisher             = Tata McGraw-Hill Education
| publication-date      = 2008
| page                  = 143
| isbn                  = 9780074637081
| access-date            = 2015-10-21
| quote                 = [...] the keyword DISTINCT [...] eliminates the duplicates from the result set.
}}
</ref>
* The {{code|OFFSET}} clause specifies the number of rows to skip before starting to return data. 
*The {{code|FETCH FIRST}} clause specifies the number of rows to return. Some SQL databases instead have non-standard alternatives, e.g. {{code|LIMIT}}, {{code|TOP}} or {{code|ROWNUM}}.
The clauses of a query have a particular order of execution,<ref>{{Cite web|url=https://www.designcise.com/web/tutorial/what-is-the-order-of-execution-of-an-sql-query|title=What Is The Order Of Execution Of An SQL Query? - Designcise.com|website=www.designcise.com|date=29 June 2015|access-date=2018-02-04}}</ref> which is denoted by the number on the right hand side. It is as follows:
{|
| <code>SELECT ''<columns>''</code> || 5.
|-
| <code>FROM ''<nowiki><table></nowiki>''</code> || 1.
|-
| <code>WHERE ''<predicate on rows>''</code> || 2.
|-
| <code>GROUP BY ''<columns>''</code> || 3.
|-
| <code>HAVING ''<predicate on groups>''</code> || 4.
|-
| <code>ORDER BY ''<columns>''</code> || 6.
|-
| <code>OFFSET</code>  || 7.
|-
| <code>FETCH FIRST</code>  || 8.
|}

The following example of a {{code|SELECT}} query 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.

<syntaxhighlight lang="sql">
SELECT *
 FROM  Book
 WHERE price > 100.00
 ORDER BY title;
</syntaxhighlight>

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.

<syntaxhighlight lang="sql">
SELECT Book.title AS Title,
       count(*) AS Authors
 FROM  Book
 JOIN  Book_author
   ON  Book.isbn = Book_author.isbn
 GROUP BY Book.title;
</syntaxhighlight>

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 ''Book'' table, one could re-write the query above in the following form:

<syntaxhighlight lang="sql">
SELECT title,
       count(*) AS Authors
 FROM  Book
 NATURAL JOIN Book_author
 GROUP BY title;
</syntaxhighlight>

However, many{{quantify|date=October 2015}} vendors either do not support this approach, or require certain column-naming conventions 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''.

<syntaxhighlight lang="sql">
SELECT isbn,
       title,
       price,
       price * 0.06 AS sales_tax
 FROM  Book
 WHERE price > 100.00
 ORDER BY title;
</syntaxhighlight>

=== Subqueries ===

Queries can be nested so that the results of one query can be used in another query via a [[relational operator]] or aggregation function. A nested query is also known as a ''subquery''. While joins and other table operations provide computationally superior (i.e. faster) alternatives in many cases, the use of subqueries introduces a hierarchy in execution that can be useful or necessary. In the following example, the aggregation function {{code|AVG}} receives as input the result of a subquery:

<syntaxhighlight lang="sql">
SELECT isbn,
       title,
       price
 FROM  Book
 WHERE price < (SELECT AVG(price) FROM Book)
 ORDER BY title;
</syntaxhighlight>

A subquery can use values from the outer query, in which case it is known as a [[correlated subquery]].

Since 1999 the SQL standard allows {{code|WITH}} clauses for subqueries, i.e. named subqueries, usually called [[common table expression]]s (also called [[subquery factoring]]). CTEs can also be [[recursive]] by referring to themselves; [[Hierarchical and recursive queries in SQL|the resulting mechanism]] allows tree or graph traversals (when represented as relations), and more generally [[fixpoint]] computations.

=== Derived table ===

A ''derived table'' is the use of referencing an SQL subquery in a FROM clause.  Essentially, the derived table is a subquery that can be selected from or joined to. The derived table functionality allows the user to reference the subquery as a table. The derived table is sometimes referred to as an ''inline view '' or a ''subselect''.

In the following example, the SQL statement involves a join from the initial "Book" table to the derived table "sales". This derived table captures associated book sales information using the ISBN to join to the "Book" table.  As a result, the derived table provides the result set with additional columns (the number of items sold and the company that sold the books):

<syntaxhighlight lang="sql">
SELECT b.isbn, b.title, b.price, sales.items_sold, sales.company_nm
FROM Book b
  JOIN (SELECT SUM(Items_Sold) Items_Sold, Company_Nm, ISBN
        FROM Book_Sales
        GROUP BY Company_Nm, ISBN) sales
  ON sales.isbn = b.isbn
</syntaxhighlight>

=== Null or three-valued logic (3VL) ===

{{main article|Null (SQL)}}

The concept of [[Null (SQL)|Null]] allows SQL to deal with missing information in the relational model. The word {{code|NULL}} is a reserved keyword in SQL, used to identify the Null special marker. Comparisons with Null, for instance equality (=) in WHERE clauses, results in an Unknown truth value. In SELECT statements SQL returns only results for which the WHERE clause returns a value of True; i.e., it excludes results with values of False and also excludes those whose value is Unknown.

Along with True and False, the Unknown resulting from direct comparisons with Null thus brings a fragment of [[three-valued logic]] to SQL. The truth tables SQL uses for AND, OR, and NOT correspond to a common fragment of the Kleene and Lukasiewicz three-valued logic (which differ in their definition of implication, however SQL defines no such operation).<ref name="Klein">{{Cite book | last1 = Hans-Joachim | first1 = K. | chapter = Null Values in Relational Databases and Sure Information Answers | doi = 10.1007/3-540-36596-6_7 | title = Semantics in Databases. Second International Workshop Dagstuhl Castle, Germany, January 7–12, 2001. Revised Papers | series = Lecture Notes in Computer Science | volume = 2582 | pages = 119–138| year = 2003 | isbn = 978-3-540-00957-3 | url = http://www.is.informatik.uni-kiel.de/~hjk/sqlni.ps| archive-url = https://web.archive.org/web/20180707170027/http://www.is.informatik.uni-kiel.de/~hjk/sqlni.ps | archive-date = 7 July 2018 }}</ref>

{| border="0"
|-
| valign="top" |
{| class="wikitable"
|-
! colspan=2 rowspan=2 | p AND q
! colspan=3 | p
|-
! style="width: 5em;" | True
! style="width: 5em;" | False
! style="width: 5em;" | Unknown
|-
! rowspan=3 style="width: 1.2em;" | q || style="width: 5em;" | True
| {{yes|True}}
| {{no|False}}
| {{Unknown}}
|-
! style="width: 5em;" | False
| {{no|False}}
| {{no|False}}
| {{no|False}}
|-
! style="width: 5em;" | Unknown
| {{Unknown}}
| {{no|False}}
| {{Unknown}}
|}
| valign="top" |
{| class="wikitable"
|-
! colspan=2 rowspan=2 | p OR q
! colspan=3 | p
|-
! style="width: 5em;" | True
! style="width: 5em;" | False
! style="width: 5em;" | Unknown
|-
! rowspan=3 style="width: 1.2em;" | q || style="width: 5em;" | True
| {{yes|True}}
| {{yes|True}}
| {{yes|True}}
|-
! style="width: 5em;" | False
| {{yes|True}}
| {{no|False}}
| {{Unknown}}
|-
! style="width: 5em;" | Unknown
| {{yes|True}}
| {{Unknown}}
| {{Unknown}}
|}
|}

{| border="0"
|-
| valign="top" |
{| class="wikitable"
|-
! colspan=2 rowspan=2 | p = q
! colspan=3 | p
|-
! style="width: 5em;" | True
! style="width: 5em;" | False
! style="width: 5em;" | Unknown
|-
! rowspan=3 style="width: 1.2em;" | q || style="width: 5em;" | True
| {{yes|True}}
| {{no|False}}
| {{Unknown}}
|-
! style="width: 5em;" | False
| {{no|False}}
| {{yes|True}}
| {{Unknown}}
|-
! style="width: 5em;" | Unknown
| {{Unknown}}
| {{Unknown}}
| {{Unknown}}
|}
| valign="top" |
{| class="wikitable"
|-
! style="width: 5em;" | q
! style="width: 5em;" | NOT q
|-
! True
| {{no|False}}
|-
! False
| {{yes|True}}
|-
! Unknown
| {{Unknown}}
|}
|}

There are however disputes about the semantic interpretation of Nulls in SQL because of its treatment outside direct comparisons. As seen in the table above, direct equality comparisons between two NULLs in SQL (e.g. {{code|NULL {{=}} NULL}}) return a truth value of Unknown. This is in line with the interpretation that 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. However, the principle that two Nulls aren't equal to each other is effectively violated in the SQL specification for the {{code|UNION}} and {{code|INTERSECT}} operators, which do identify nulls with each other.<ref name="Meyden">[https://books.google.com/books?id=gF0b85IuqQwC&pg=PA344 Ron van der Meyden, "Logical approaches to incomplete information: a survey" in Chomicki, Jan; Saake, Gunter (Eds.) ''Logics for Databases and Information Systems'', Kluwer Academic Publishers] {{ISBN|978-0-7923-8129-7}}, p. 344</ref> Consequently, these [[Set operations (SQL)|set operations in SQL]] may produce results not representing sure information, unlike operations involving explicit comparisons with NULL (e.g. those in a {{code|WHERE}} clause discussed above). In Codd's 1979 proposal (which was basically adopted by SQL92) this semantic inconsistency is rationalized by arguing that removal of duplicates in set operations happens "at a lower level of detail than equality testing in the evaluation of retrieval operations".<ref name="Klein" /> However, computer-science professor Ron van der Meyden concluded that "The inconsistencies in the SQL standard mean that it is not possible to ascribe any intuitive logical semantics to the treatment of nulls in SQL."<ref name="Meyden" />

Additionally, because SQL operators return Unknown when comparing anything with Null directly, SQL provides two Null-specific comparison predicates: {{code|IS NULL}} and {{code|IS NOT NULL}} test whether data is or is not Null.<ref name="SQL2003-Part2-Sec87">{{cite book | last = ISO/IEC | title = ISO/IEC 9075-2:2003, "SQL/Foundation" | publisher = ISO/IEC }}</ref> SQL does not explicitly support [[universal quantification]], and must work it out as a negated [[existential quantification]].<ref name="Negri">{{cite journal|title=Semantics and problems of universal quantification in SQL |doi=10.1093/comjnl/32.1.90 |journal=The Computer Journal |date= February 1989|volume=32 |issue=1 |pages=90–91 |last1=Negri |first1=M. |last2=Pelagatti |first2=G. |last3=Sbattella |first3=L. |doi-access=free }}</ref><ref name="Fratarcangeli">{{cite journal|url=https://dl.acm.org/doi/10.1145/126482.126484 |title=Technique for universal quantification in SQL |doi=10.1145/126482.126484 |journal=ACM SIGMOD Record |year = 1991|access-date=2024-11-07|last1 = Fratarcangeli|first1 = Claudio|volume = 20|issue = 3|pages = 16–24|s2cid=18326990 |doi-access=free }}</ref><ref name="kawash">[https://www.thefreelibrary.com/Complex+quantification+in+Structured+Query+Language+(SQL):+a+tutorial...-a0119901477 Kawash, Jalal (2004) Complex quantification in Structured Query Language (SQL): a tutorial using relational calculus]; ''Journal of Computers in Mathematics and Science Teaching'' {{ISSN|0731-9258}} Volume 23, Issue 2, 2004 AACE Norfolk, Virginia. Thefreelibrary.com</ref> There is also the {{code|<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 {{code|NOT (<row value expression> IS DISTINCT FROM <row value expression>)}}. [[SQL:1999]] also introduced {{code|BOOLEAN}} type variables, which according to the standard can also hold Unknown values if it is nullable. In practice, a number of systems (e.g. [[PostgreSQL]]) implement the BOOLEAN Unknown as a BOOLEAN NULL, which the standard says that the NULL BOOLEAN and UNKNOWN "may be used interchangeably to mean exactly the same thing".<ref>{{cite book|author=C. Date|title=SQL and Relational Theory: How to Write Accurate SQL Code|url=https://books.google.com/books?id=Ew06OZtjuJEC&pg=PA83|year=2011|publisher=O'Reilly Media, Inc.|isbn=978-1-4493-1640-2|page=83}}</ref><ref>ISO/IEC 9075-2:2011 §4.5</ref>

=== Data manipulation ===

The [[Data Manipulation Language]] (DML) is the subset of SQL used to add, update and delete data:
* <code>[[Insert (SQL)|INSERT]]</code> adds rows (formally [[tuple]]s) to an existing table, e.g.:

<syntaxhighlight lang="sql">
INSERT INTO example
 (column1, column2, column3)
 VALUES
 ('test', 'N', NULL);
</syntaxhighlight>

* <code>[[Update (SQL)|UPDATE]]</code> modifies a set of existing table rows, e.g.:

<syntaxhighlight lang="sql">
UPDATE example
 SET column1 = 'updated value'
 WHERE column2 = 'N';
</syntaxhighlight>

* <code>[[Delete (SQL)|DELETE]]</code> removes existing rows from a table, e.g.:

<syntaxhighlight lang="sql">
DELETE FROM example
 WHERE column2 = 'N';
</syntaxhighlight>

* <code>[[Merge (SQL)|MERGE]]</code> is used to combine the data of multiple tables. It combines the {{code|INSERT}} and {{code|UPDATE}} elements. It is defined in the SQL:2003 standard; prior to that, some databases provided similar functionality via different syntax, sometimes called "[[upsert]]".

<syntaxhighlight lang="sql">
 MERGE INTO table_name USING table_reference ON (condition)
 WHEN MATCHED THEN
 UPDATE SET column1 = value1 [, column2 = value2 ...]
 WHEN NOT MATCHED THEN
 INSERT (column1 [, column2 ...]) VALUES (value1 [, value2 ...])
</syntaxhighlight>

=== Transaction controls ===

Transactions, if available, wrap DML operations:

* {{code|START TRANSACTION}} (or {{code|BEGIN WORK}}, or {{code|BEGIN TRANSACTION}}, depending on SQL dialect) marks the start of a [[database transaction]], which either completes entirely or not at all.
* {{code|SAVE TRANSACTION}} (or {{code|SAVEPOINT}}) saves the state of the database at the current point in transaction

<syntaxhighlight lang="postgresql">
CREATE TABLE tbl_1(id int);
 INSERT INTO tbl_1(id) VALUES(1);
 INSERT INTO tbl_1(id) VALUES(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;
</syntaxhighlight>

* <code>[[Commit (SQL)|COMMIT]]</code> makes all data changes in a transaction permanent.
* <code>[[Rollback (SQL)|ROLLBACK]]</code> discards all data changes since the last {{code|COMMIT}} or {{code|ROLLBACK}}, leaving the data as it was prior to those changes. Once the {{code|COMMIT}} statement completes, the transaction's changes cannot be rolled back.

{{code|COMMIT}} and {{code|ROLLBACK}} terminate the current transaction and release data locks. In the absence of a {{code|START TRANSACTION}} or similar statement, the semantics of SQL are implementation-dependent.
The following example shows a classic transfer of funds transaction, where money is removed from one account and added to another. If either the removal or the addition fails, the entire transaction is rolled back.

<syntaxhighlight lang="plpgsql">
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;
</syntaxhighlight>

=== Data definition ===

The [[Data Definition Language]] (DDL) manages table and index structure. The most basic items of DDL are the {{code|CREATE}}, {{code|ALTER}}, {{code|RENAME}}, {{code|DROP}} and {{code|TRUNCATE}} statements:

* <code>[[Create (SQL)|CREATE]]</code> creates an object (a table, for example) in the database, e.g.:

<syntaxhighlight lang="sql">
CREATE TABLE example(
 column1 INTEGER,
 column2 VARCHAR(50),
 column3 DATE NOT NULL,
 PRIMARY KEY (column1, column2)
);
</syntaxhighlight>

* <code>[[Alter (SQL)|ALTER]]</code> modifies the structure of an existing object in various ways, for example, adding a column to an existing table or a constraint, e.g.:

<syntaxhighlight lang="sql">
ALTER TABLE example ADD column4 INTEGER DEFAULT 25 NOT NULL;
</syntaxhighlight>

* <code>[[Truncate (SQL)|TRUNCATE]]</code> deletes 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 (data is not written to the logs for rollback later, unlike DELETE).

<syntaxhighlight lang="sql">
TRUNCATE TABLE example;
</syntaxhighlight>

* <code>[[Drop (SQL)|DROP]]</code> deletes an object in the database, usually irretrievably, i.e., it cannot be rolled back, e.g.:

<syntaxhighlight lang="sql">
DROP TABLE example;
</syntaxhighlight>

=== Data types ===

Each column in an SQL table declares the type(s) that column may contain. ANSI SQL includes the following data types.<ref name="ISO/IEC">{{cite web | url = https://www.iso.org/standard/63555.html | title = ISO/IEC 9075-1:2016: Information technology&nbsp;– Database languages&nbsp;– SQL&nbsp;– Part 1: Framework (SQL/Framework) }}</ref>

; Character strings and national character strings
* <code>CHARACTER(<var>n</var>)</code> (or <code>CHAR(<var>n</var>)</code>): fixed-width <var>n</var>-character string, padded with spaces as needed
* <code>CHARACTER VARYING(<var>n</var>)</code> (or <code>VARCHAR(<var>n</var>)</code>): variable-width string with a maximum size of <var>n</var> characters
* <code>CHARACTER LARGE OBJECT(<var>n</var> [ K | M | G | T ])</code> (or <code>CLOB(<var>n</var> [ K | M | G | T ])</code>): character large object with a maximum size of <var>n [ K | M | G | T ]</var> characters
* <code>NATIONAL CHARACTER(<var>n</var>)</code> (or <code>NCHAR(<var>n</var>)</code>): fixed width string supporting an international character set
* <code>NATIONAL CHARACTER VARYING(<var>n</var>)</code> (or <code>NVARCHAR(<var>n</var>)</code>): variable-width <code>NCHAR</code> string
* <code>NATIONAL CHARACTER LARGE OBJECT(<var>n</var> [ K | M | G | T ])</code> (or <code>NCLOB(<var>n</var> [ K | M | G | T ])</code>): national character large object with a maximum size of <var>n [ K | M | G | T ]</var> characters

For the {{code|CHARACTER LARGE OBJECT}} and {{code|NATIONAL CHARACTER LARGE OBJECT}} data types, the multipliers {{code|K}} (1 024), {{code|M}} (1 048 576), {{code|G}} (1 073 741 824) and {{code|T}} (1 099 511 627 776) can be optionally used when specifying the length.

; Binary
* <code>BINARY(<var>n</var>)</code>: Fixed length binary string, maximum length <var>n</var>.
* <code>BINARY VARYING(<var>n</var>)</code> (or <code>VARBINARY(<var>n</var>)</code>): Variable length binary string, maximum length <var>n</var>.
* <code>BINARY LARGE OBJECT(<var>n</var> [ K | M | G | T ])</code> (or <code>BLOB(<var>n</var> [ K | M | G | T ])</code>): binary large object with a maximum length <var>n [ K | M | G | T ]</var>.

For the {{code|BINARY LARGE OBJECT}} data type, the multipliers {{code|K}} (1 024), {{code|M}} (1 048 576), {{code|G}} (1 073 741 824) and {{code|T}} (1 099 511 627 776) can be optionally used when specifying the length.

; Boolean
* {{code|BOOLEAN}}

The {{code|BOOLEAN}} data type can store the values {{code|TRUE}} and {{code|FALSE}}.
 
; Numerical
* {{code|INTEGER}} (or {{code|INT}}), {{code|SMALLINT}} and {{code|BIGINT}}
* {{code|FLOAT}}, {{code|REAL}} and {{code|DOUBLE PRECISION}}
* <code>NUMERIC(<var>precision</var>, <var>scale</var>)</code> or <code>DECIMAL(<var>precision</var>, <var>scale</var>)</code>
* <code>DECFLOAT(<var>precision</var></code>)

For example, the number 123.45 has a precision of 5 and a scale of 2. The <var>precision</var> is a positive integer that determines the number of significant digits in a particular radix (binary or decimal). The <var>scale</var> is a non-negative integer. A scale of 0 indicates that the number is an integer. For a decimal number with scale S, the exact numeric value is the integer value of the significant digits divided by 10<sup>S</sup>.

SQL provides the functions {{code|CEILING}} and {{code|FLOOR}} to round numerical values. (Popular vendor specific functions are {{code|TRUNC}} (Informix, DB2, PostgreSQL, Oracle and MySQL) and {{code|ROUND}} (Informix, SQLite, Sybase, Oracle, PostgreSQL, Microsoft SQL Server and Mimer SQL.))

; Temporal (datetime)
* {{code|DATE}}: for date values (e.g. {{code|2011-05-03}}).
* {{code|TIME}}: for time values (e.g. {{code|15:51:36}}).
* {{code|TIME WITH TIME ZONE}}: the same as {{code|TIME}}, but including details about the time zone in question.
* {{code|TIMESTAMP}}: This is a {{code|DATE}} and a {{code|TIME}} put together in one variable (e.g. {{code|2011-05-03 15:51:36.123456}}).
* {{code|TIMESTAMP WITH TIME ZONE}}: the same as {{code|TIMESTAMP}}, but including details about the time zone in question.

The SQL function {{code|EXTRACT}} can be used for extracting a single field (seconds, for instance) of a datetime or interval value. The current system date / time of the database server can be called by using functions like {{code|CURRENT_DATE}}, {{code|CURRENT_TIMESTAMP}}, {{code|LOCALTIME}}, or {{code|LOCALTIMESTAMP}}. (Popular vendor specific functions are {{code|TO_DATE}}, {{code|TO_TIME}}, {{code|TO_TIMESTAMP}}, {{code|YEAR}}, {{code|MONTH}}, {{code|DAY}}, {{code|HOUR}}, {{code|MINUTE}}, {{code|SECOND}}, {{code|DAYOFYEAR}}, {{code|DAYOFMONTH}} and {{code|DAYOFWEEK}}.)

; Interval (datetime)
* <code>YEAR(<var>precision</var>)</code>: a number of years
* <code>YEAR(<var>precision</var>) TO MONTH</code>: a number of years and months
* <code>MONTH(<var>precision</var>)</code>: a number of months
* <code>DAY(<var>precision</var>)</code>: a number of days
* <code>DAY(<var>precision</var>) TO HOUR</code>: a number of days and hours
* <code>DAY(<var>precision</var>) TO MINUTE</code>: a number of days, hours and minutes
* <code>DAY(<var>precision</var>) TO SECOND(<var>scale</var>)</code>: a number of days, hours, minutes and seconds
* <code>HOUR(<var>precision</var>)</code>: a number of hours
* <code>HOUR(<var>precision</var>) TO MINUTE</code>: a number of hours and minutes
* <code>HOUR(<var>precision</var>) TO SECOND(<var>scale</var>)</code>: a number of hours, minutes and seconds
* <code>MINUTE(<var>precision</var>)</code>: a number of minutes
* <code>MINUTE(<var>precision</var>) TO SECOND(<var>scale</var>)</code>: a number of minutes and seconds

=== Data control ===

The [[Data Control Language]] (DCL) authorizes users to access and manipulate data.
Its two main statements are:

* {{code|GRANT}} authorizes one or more users to perform an operation or a set of operations on an object.
* {{code|REVOKE}} eliminates a grant, which may be the default grant.

Example:

<syntaxhighlight lang="sql">
GRANT SELECT, UPDATE
 ON example
 TO some_user, another_user;

REVOKE SELECT, UPDATE
 ON example
 FROM some_user, another_user;
</syntaxhighlight>

==References==
{{Reflist}}

==External links==
{{Sister project links
|commons=Category:Structured Query Language
|v=Introduction to SQL
|b=Structured Query Language
|n=no|s=no|q=no}}
* {{Citation |url=https://standards.iso.org/iso-iec/9075/-2/ed-6/en/ISO_IEC_9075-2(E)_Foundation.bnf.xml |contribution=Definition of SQL grammar |title=ISO_IEC_9075-2(E)_Foundation.bnf.xml |publisher=ISO Standards Maintenance Portal |format=[[Backus–Naur form]] XML}}.
* {{Citation |url=https://quickformatter.com/sql-formatter |contribution=Online SQL Formating Utility |title=Online SQL Formatter Utility |archive-url=https://web.archive.org/web/20240918172914/https://quickformatter.com/sql-formatter |archive-date=2024-09-18 |url-status=dead}}.

{{SQL}}
{{Databases}}
{{Query languages}}
{{ISO standards}}

[[Category:SQL]]
[[Category:Query languages]]
[[Category:Programming language syntax]]
[[Category:Declarative programming languages]]
[[Category:Articles with example SQL code]]
[[Category:Data modeling languages]]
[[Category:Relational database management systems]]
[[Category:Database management systems]]