<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<HTML
><HEAD
><TITLE
>SQL Syntax</TITLE
><META
NAME="GENERATOR"
CONTENT="Modular DocBook HTML Stylesheet Version 1.7"><LINK
REV="MADE"
HREF="mailto:pgsql-docs@postgresql.org"><LINK
REL="HOME"
TITLE="PostgreSQL 7.4.1 Documentation"
HREF="index.html"><LINK
REL="UP"
TITLE="The SQL Language"
HREF="sql.html"><LINK
REL="PREVIOUS"
TITLE="The SQL Language"
HREF="sql.html"><LINK
REL="NEXT"
TITLE="Value Expressions"
HREF="sql-expressions.html"><LINK
REL="STYLESHEET"
TYPE="text/css"
HREF="stylesheet.css"><META
NAME="creation"
CONTENT="2003-12-22T03:48:47"></HEAD
><BODY
CLASS="CHAPTER"
><DIV
CLASS="NAVHEADER"
><TABLE
SUMMARY="Header navigation table"
WIDTH="100%"
BORDER="0"
CELLPADDING="0"
CELLSPACING="0"
><TR
><TH
COLSPAN="5"
ALIGN="center"
VALIGN="bottom"
>PostgreSQL 7.4.1 Documentation</TH
></TR
><TR
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="top"
><A
HREF="sql.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="10%"
ALIGN="left"
VALIGN="top"
><A
HREF="sql.html#AEN997"
>Fast Backward</A
></TD
><TD
WIDTH="60%"
ALIGN="center"
VALIGN="bottom"
></TD
><TD
WIDTH="10%"
ALIGN="right"
VALIGN="top"
><A
HREF="ddl.html"
>Fast Forward</A
></TD
><TD
WIDTH="10%"
ALIGN="right"
VALIGN="top"
><A
HREF="sql-expressions.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
></TABLE
><HR
ALIGN="LEFT"
WIDTH="100%"></DIV
><DIV
CLASS="CHAPTER"
><H1
><A
NAME="SQL-SYNTAX"
></A
>Chapter 4. SQL Syntax</H1
><DIV
CLASS="TOC"
><DL
><DT
><B
>Table of Contents</B
></DT
><DT
>4.1. <A
HREF="sql-syntax.html#SQL-SYNTAX-LEXICAL"
>Lexical Structure</A
></DT
><DD
><DL
><DT
>4.1.1. <A
HREF="sql-syntax.html#SQL-SYNTAX-IDENTIFIERS"
>Identifiers and Key Words</A
></DT
><DT
>4.1.2. <A
HREF="sql-syntax.html#SQL-SYNTAX-CONSTANTS"
>Constants</A
></DT
><DT
>4.1.3. <A
HREF="sql-syntax.html#SQL-SYNTAX-OPERATORS"
>Operators</A
></DT
><DT
>4.1.4. <A
HREF="sql-syntax.html#AEN1269"
>Special Characters</A
></DT
><DT
>4.1.5. <A
HREF="sql-syntax.html#SQL-SYNTAX-COMMENTS"
>Comments</A
></DT
><DT
>4.1.6. <A
HREF="sql-syntax.html#SQL-PRECEDENCE"
>Lexical Precedence</A
></DT
></DL
></DD
><DT
>4.2. <A
HREF="sql-expressions.html"
>Value Expressions</A
></DT
><DD
><DL
><DT
>4.2.1. <A
HREF="sql-expressions.html#AEN1510"
>Column References</A
></DT
><DT
>4.2.2. <A
HREF="sql-expressions.html#AEN1526"
>Positional Parameters</A
></DT
><DT
>4.2.3. <A
HREF="sql-expressions.html#AEN1540"
>Subscripts</A
></DT
><DT
>4.2.4. <A
HREF="sql-expressions.html#AEN1559"
>Field Selection</A
></DT
><DT
>4.2.5. <A
HREF="sql-expressions.html#AEN1570"
>Operator Invocations</A
></DT
><DT
>4.2.6. <A
HREF="sql-expressions.html#AEN1599"
>Function Calls</A
></DT
><DT
>4.2.7. <A
HREF="sql-expressions.html#SYNTAX-AGGREGATES"
>Aggregate Expressions</A
></DT
><DT
>4.2.8. <A
HREF="sql-expressions.html#SQL-SYNTAX-TYPE-CASTS"
>Type Casts</A
></DT
><DT
>4.2.9. <A
HREF="sql-expressions.html#SQL-SYNTAX-SCALAR-SUBQUERIES"
>Scalar Subqueries</A
></DT
><DT
>4.2.10. <A
HREF="sql-expressions.html#SQL-SYNTAX-ARRAY-CONSTRUCTORS"
>Array Constructors</A
></DT
><DT
>4.2.11. <A
HREF="sql-expressions.html#SYNTAX-EXPRESS-EVAL"
>Expression Evaluation Rules</A
></DT
></DL
></DD
></DL
></DIV
><A
NAME="AEN1014"
></A
><P
> This chapter describes the syntax of SQL. It forms the foundation
for understanding the following chapters which will go into detail
about how the SQL commands are applied to define and modify data.
</P
><P
> We also advise users who are already familiar with SQL to read this
chapter carefully because there are several rules and concepts that
are implemented inconsistently among SQL databases or that are
specific to <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
>.
</P
><DIV
CLASS="SECT1"
><H1
CLASS="SECT1"
><A
NAME="SQL-SYNTAX-LEXICAL"
>4.1. Lexical Structure</A
></H1
><A
NAME="AEN1022"
></A
><P
> SQL input consists of a sequence of
<I
CLASS="FIRSTTERM"
>commands</I
>. A command is composed of a
sequence of <I
CLASS="FIRSTTERM"
>tokens</I
>, terminated by a
semicolon (<SPAN
CLASS="QUOTE"
>";"</SPAN
>). The end of the input stream also
terminates a command. Which tokens are valid depends on the syntax
of the particular command.
</P
><P
> A token can be a <I
CLASS="FIRSTTERM"
>key word</I
>, an
<I
CLASS="FIRSTTERM"
>identifier</I
>, a <I
CLASS="FIRSTTERM"
>quoted
identifier</I
>, a <I
CLASS="FIRSTTERM"
>literal</I
> (or
constant), or a special character symbol. Tokens are normally
separated by whitespace (space, tab, newline), but need not be if
there is no ambiguity (which is generally only the case if a
special character is adjacent to some other token type).
</P
><P
> Additionally, <I
CLASS="FIRSTTERM"
>comments</I
> can occur in SQL
input. They are not tokens, they are effectively equivalent to
whitespace.
</P
><P
> For example, the following is (syntactically) valid SQL input:
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT * FROM MY_TABLE;
UPDATE MY_TABLE SET A = 5;
INSERT INTO MY_TABLE VALUES (3, 'hi there');</PRE
><P>
This is a sequence of three commands, one per line (although this
is not required; more than one command can be on a line, and
commands can usefully be split across lines).
</P
><P
> The SQL syntax is not very consistent regarding what tokens
identify commands and which are operands or parameters. The first
few tokens are generally the command name, so in the above example
we would usually speak of a <SPAN
CLASS="QUOTE"
>"SELECT"</SPAN
>, an
<SPAN
CLASS="QUOTE"
>"UPDATE"</SPAN
>, and an <SPAN
CLASS="QUOTE"
>"INSERT"</SPAN
> command. But
for instance the <TT
CLASS="COMMAND"
>UPDATE</TT
> command always requires
a <TT
CLASS="TOKEN"
>SET</TT
> token to appear in a certain position, and
this particular variation of <TT
CLASS="COMMAND"
>INSERT</TT
> also
requires a <TT
CLASS="TOKEN"
>VALUES</TT
> in order to be complete. The
precise syntax rules for each command are described in <A
HREF="reference.html"
>Part VI</A
>.
</P
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="SQL-SYNTAX-IDENTIFIERS"
>4.1.1. Identifiers and Key Words</A
></H2
><A
NAME="AEN1048"
></A
><A
NAME="AEN1051"
></A
><A
NAME="AEN1054"
></A
><P
> Tokens such as <TT
CLASS="TOKEN"
>SELECT</TT
>, <TT
CLASS="TOKEN"
>UPDATE</TT
>, or
<TT
CLASS="TOKEN"
>VALUES</TT
> in the example above are examples of
<I
CLASS="FIRSTTERM"
>key words</I
>, that is, words that have a fixed
meaning in the SQL language. The tokens <TT
CLASS="TOKEN"
>MY_TABLE</TT
>
and <TT
CLASS="TOKEN"
>A</TT
> are examples of
<I
CLASS="FIRSTTERM"
>identifiers</I
>. They identify names of
tables, columns, or other database objects, depending on the
command they are used in. Therefore they are sometimes simply
called <SPAN
CLASS="QUOTE"
>"names"</SPAN
>. Key words and identifiers have the
same lexical structure, meaning that one cannot know whether a
token is an identifier or a key word without knowing the language.
A complete list of key words can be found in <A
HREF="sql-keywords-appendix.html"
>Appendix C</A
>.
</P
><P
> SQL identifiers and key words must begin with a letter
(<TT
CLASS="LITERAL"
>a</TT
>-<TT
CLASS="LITERAL"
>z</TT
>, but also letters with
diacritical marks and non-Latin letters) or an underscore
(<TT
CLASS="LITERAL"
>_</TT
>). Subsequent characters in an identifier or
key word can be letters, underscores, digits
(<TT
CLASS="LITERAL"
>0</TT
>-<TT
CLASS="LITERAL"
>9</TT
>), or dollar signs
(<TT
CLASS="LITERAL"
>$</TT
>). Note that dollar signs are not allowed in identifiers
according to the letter of the SQL standard, so their use may render
applications less portable.
The SQL standard will not define a key word that contains
digits or starts or ends with an underscore, so identifiers of this
form are safe against possible conflict with future extensions of the
standard.
</P
><P
> <A
NAME="AEN1075"
></A
>
The system uses no more than <TT
CLASS="SYMBOL"
>NAMEDATALEN</TT
>-1
characters of an identifier; longer names can be written in
commands, but they will be truncated. By default,
<TT
CLASS="SYMBOL"
>NAMEDATALEN</TT
> is 64 so the maximum identifier
length is 63. If this limit is problematic, it can be raised by
changing the <TT
CLASS="SYMBOL"
>NAMEDATALEN</TT
> constant in
<TT
CLASS="FILENAME"
>src/include/postgres_ext.h</TT
>.
</P
><P
> <A
NAME="AEN1083"
></A
>
Identifier and key word names are case insensitive. Therefore
</P><PRE
CLASS="PROGRAMLISTING"
>UPDATE MY_TABLE SET A = 5;</PRE
><P>
can equivalently be written as
</P><PRE
CLASS="PROGRAMLISTING"
>uPDaTE my_TabLE SeT a = 5;</PRE
><P>
A convention often used is to write key words in upper
case and names in lower case, e.g.,
</P><PRE
CLASS="PROGRAMLISTING"
>UPDATE my_table SET a = 5;</PRE
><P>
</P
><P
> <A
NAME="AEN1090"
></A
>
There is a second kind of identifier: the <I
CLASS="FIRSTTERM"
>delimited
identifier</I
> or <I
CLASS="FIRSTTERM"
>quoted
identifier</I
>. It is formed by enclosing an arbitrary
sequence of characters in double-quotes
(<TT
CLASS="LITERAL"
>"</TT
>). A delimited
identifier is always an identifier, never a key word. So
<TT
CLASS="LITERAL"
>"select"</TT
> could be used to refer to a column or
table named <SPAN
CLASS="QUOTE"
>"select"</SPAN
>, whereas an unquoted
<TT
CLASS="LITERAL"
>select</TT
> would be taken as a key word and
would therefore provoke a parse error when used where a table or
column name is expected. The example can be written with quoted
identifiers like this:
</P><PRE
CLASS="PROGRAMLISTING"
>UPDATE "my_table" SET "a" = 5;</PRE
><P>
</P
><P
> Quoted identifiers can contain any character other than a double
quote itself. (To include a double quote, write two double quotes.)
This allows constructing table or column names that would
otherwise not be possible, such as ones containing spaces or
ampersands. The length limitation still applies.
</P
><P
> Quoting an identifier also makes it case-sensitive, whereas
unquoted names are always folded to lower case. For example, the
identifiers <TT
CLASS="LITERAL"
>FOO</TT
>, <TT
CLASS="LITERAL"
>foo</TT
>, and
<TT
CLASS="LITERAL"
>"foo"</TT
> are considered the same by
<SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
>, but
<TT
CLASS="LITERAL"
>"Foo"</TT
> and <TT
CLASS="LITERAL"
>"FOO"</TT
> are
different from these three and each other. (The folding of
unquoted names to lower case in <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> is
incompatible with the SQL standard, which says that unquoted names
should be folded to upper case. Thus, <TT
CLASS="LITERAL"
>foo</TT
>
should be equivalent to <TT
CLASS="LITERAL"
>"FOO"</TT
> not
<TT
CLASS="LITERAL"
>"foo"</TT
> according to the standard. If you want
to write portable applications you are advised to always quote a
particular name or never quote it.)
</P
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="SQL-SYNTAX-CONSTANTS"
>4.1.2. Constants</A
></H2
><A
NAME="AEN1114"
></A
><P
> There are three kinds of <I
CLASS="FIRSTTERM"
>implicitly-typed
constants</I
> in <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
>:
strings, bit strings, and numbers.
Constants can also be specified with explicit types, which can
enable more accurate representation and more efficient handling by
the system. The implicit constants are described below; explicit
constants are discussed afterwards.
</P
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="SQL-SYNTAX-STRINGS"
>4.1.2.1. String Constants</A
></H3
><A
NAME="AEN1121"
></A
><P
> <A
NAME="AEN1125"
></A
>
A string constant in SQL is an arbitrary sequence of characters
bounded by single quotes (<TT
CLASS="LITERAL"
>'</TT
>), e.g., <TT
CLASS="LITERAL"
>'This
is a string'</TT
>. SQL allows single quotes to be embedded
in strings by typing two adjacent single quotes, e.g.,
<TT
CLASS="LITERAL"
>'Dianne''s horse'</TT
>. In
<SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> single quotes may
alternatively be escaped with a backslash (<TT
CLASS="LITERAL"
>\</TT
>),
e.g., <TT
CLASS="LITERAL"
>'Dianne\'s horse'</TT
>.
</P
><P
> C-style backslash escapes are also available:
<TT
CLASS="LITERAL"
>\b</TT
> is a backspace, <TT
CLASS="LITERAL"
>\f</TT
> is a
form feed, <TT
CLASS="LITERAL"
>\n</TT
> is a newline,
<TT
CLASS="LITERAL"
>\r</TT
> is a carriage return, <TT
CLASS="LITERAL"
>\t</TT
>
is a tab, and <TT
CLASS="LITERAL"
>\<VAR
CLASS="REPLACEABLE"
>xxx</VAR
></TT
>,
where <VAR
CLASS="REPLACEABLE"
>xxx</VAR
> is an octal number, is a
byte with the corresponding code. (It is your responsibility
that the byte sequences you create are valid characters in the
server character set encoding.) Any other character following a
backslash is taken literally. Thus, to include a backslash in a
string constant, type two backslashes.
</P
><P
> The character with the code zero cannot be in a string constant.
</P
><P
> Two string constants that are only separated by whitespace
<SPAN
CLASS="emphasis"
><I
CLASS="EMPHASIS"
>with at least one newline</I
></SPAN
> are concatenated
and effectively treated as if the string had been written in one
constant. For example:
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT 'foo'
'bar';</PRE
><P>
is equivalent to
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT 'foobar';</PRE
><P>
but
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT 'foo' 'bar';</PRE
><P>
is not valid syntax. (This slightly bizarre behavior is specified
by <ACRONYM
CLASS="ACRONYM"
>SQL</ACRONYM
>; <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> is
following the standard.)
</P
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="SQL-SYNTAX-BIT-STRINGS"
>4.1.2.2. Bit-String Constants</A
></H3
><A
NAME="AEN1153"
></A
><P
> Bit-string constants look like string constants with a
<TT
CLASS="LITERAL"
>B</TT
> (upper or lower case) immediately before the
opening quote (no intervening whitespace), e.g.,
<TT
CLASS="LITERAL"
>B'1001'</TT
>. The only characters allowed within
bit-string constants are <TT
CLASS="LITERAL"
>0</TT
> and
<TT
CLASS="LITERAL"
>1</TT
>.
</P
><P
> Alternatively, bit-string constants can be specified in hexadecimal
notation, using a leading <TT
CLASS="LITERAL"
>X</TT
> (upper or lower case),
e.g., <TT
CLASS="LITERAL"
>X'1FF'</TT
>. This notation is equivalent to
a bit-string constant with four binary digits for each hexadecimal digit.
</P
><P
> Both forms of bit-string constant can be continued
across lines in the same way as regular string constants.
</P
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="AEN1165"
>4.1.2.3. Numeric Constants</A
></H3
><A
NAME="AEN1167"
></A
><P
> Numeric constants are accepted in these general forms:
</P><PRE
CLASS="SYNOPSIS"
><VAR
CLASS="REPLACEABLE"
>digits</VAR
>
<VAR
CLASS="REPLACEABLE"
>digits</VAR
>.[<SPAN
CLASS="OPTIONAL"
><VAR
CLASS="REPLACEABLE"
>digits</VAR
></SPAN
>][<SPAN
CLASS="OPTIONAL"
>e[<SPAN
CLASS="OPTIONAL"
>+-</SPAN
>]<VAR
CLASS="REPLACEABLE"
>digits</VAR
></SPAN
>]
[<SPAN
CLASS="OPTIONAL"
><VAR
CLASS="REPLACEABLE"
>digits</VAR
></SPAN
>].<VAR
CLASS="REPLACEABLE"
>digits</VAR
>[<SPAN
CLASS="OPTIONAL"
>e[<SPAN
CLASS="OPTIONAL"
>+-</SPAN
>]<VAR
CLASS="REPLACEABLE"
>digits</VAR
></SPAN
>]
<VAR
CLASS="REPLACEABLE"
>digits</VAR
>e[<SPAN
CLASS="OPTIONAL"
>+-</SPAN
>]<VAR
CLASS="REPLACEABLE"
>digits</VAR
></PRE
><P>
where <VAR
CLASS="REPLACEABLE"
>digits</VAR
> is one or more decimal
digits (0 through 9). At least one digit must be before or after the
decimal point, if one is used. At least one digit must follow the
exponent marker (<TT
CLASS="LITERAL"
>e</TT
>), if one is present.
There may not be any spaces or other characters embedded in the
constant. Note that any leading plus or minus sign is not actually
considered part of the constant; it is an operator applied to the
constant.
</P
><P
> These are some examples of valid numeric constants:
<P
CLASS="LITERALLAYOUT"
>42<br>
3.5<br>
4.<br>
.001<br>
5e2<br>
1.925e-3</P
>
</P
><P
> <A
NAME="AEN1193"
></A
>
<A
NAME="AEN1195"
></A
>
<A
NAME="AEN1197"
></A
>
A numeric constant that contains neither a decimal point nor an
exponent is initially presumed to be type <TT
CLASS="TYPE"
>integer</TT
> if its
value fits in type <TT
CLASS="TYPE"
>integer</TT
> (32 bits); otherwise it is
presumed to be type <TT
CLASS="TYPE"
>bigint</TT
> if its
value fits in type <TT
CLASS="TYPE"
>bigint</TT
> (64 bits); otherwise it is
taken to be type <TT
CLASS="TYPE"
>numeric</TT
>. Constants that contain decimal
points and/or exponents are always initially presumed to be type
<TT
CLASS="TYPE"
>numeric</TT
>.
</P
><P
> The initially assigned data type of a numeric constant is just a
starting point for the type resolution algorithms. In most cases
the constant will be automatically coerced to the most
appropriate type depending on context. When necessary, you can
force a numeric value to be interpreted as a specific data type
by casting it.<A
NAME="AEN1206"
></A
>
For example, you can force a numeric value to be treated as type
<TT
CLASS="TYPE"
>real</TT
> (<TT
CLASS="TYPE"
>float4</TT
>) by writing
</P><PRE
CLASS="PROGRAMLISTING"
>REAL '1.23' -- string style
1.23::REAL -- PostgreSQL (historical) style</PRE
><P>
</P
></DIV
><DIV
CLASS="SECT3"
><H3
CLASS="SECT3"
><A
NAME="SQL-SYNTAX-CONSTANTS-GENERIC"
>4.1.2.4. Constants of Other Types</A
></H3
><A
NAME="AEN1213"
></A
><P
> A constant of an <SPAN
CLASS="emphasis"
><I
CLASS="EMPHASIS"
>arbitrary</I
></SPAN
> type can be
entered using any one of the following notations:
</P><PRE
CLASS="SYNOPSIS"
><VAR
CLASS="REPLACEABLE"
>type</VAR
> '<VAR
CLASS="REPLACEABLE"
>string</VAR
>'
'<VAR
CLASS="REPLACEABLE"
>string</VAR
>'::<VAR
CLASS="REPLACEABLE"
>type</VAR
>
CAST ( '<VAR
CLASS="REPLACEABLE"
>string</VAR
>' AS <VAR
CLASS="REPLACEABLE"
>type</VAR
> )</PRE
><P>
The string's text is passed to the input conversion
routine for the type called <VAR
CLASS="REPLACEABLE"
>type</VAR
>. The
result is a constant of the indicated type. The explicit type
cast may be omitted if there is no ambiguity as to the type the
constant must be (for example, when it is passed as an argument
to a non-overloaded function), in which case it is automatically
coerced.
</P
><P
> It is also possible to specify a type coercion using a function-like
syntax:
</P><PRE
CLASS="SYNOPSIS"
><VAR
CLASS="REPLACEABLE"
>typename</VAR
> ( '<VAR
CLASS="REPLACEABLE"
>string</VAR
>' )</PRE
><P>
but not all type names may be used in this way; see <A
HREF="sql-expressions.html#SQL-SYNTAX-TYPE-CASTS"
>Section 4.2.8</A
> for details.
</P
><P
> The <TT
CLASS="LITERAL"
>::</TT
>, <TT
CLASS="LITERAL"
>CAST()</TT
>, and
function-call syntaxes can also be used to specify run-time type
conversions of arbitrary expressions, as discussed in <A
HREF="sql-expressions.html#SQL-SYNTAX-TYPE-CASTS"
>Section 4.2.8</A
>. But the form
<TT
CLASS="LITERAL"
><VAR
CLASS="REPLACEABLE"
>type</VAR
> '<VAR
CLASS="REPLACEABLE"
>string</VAR
>'</TT
>
can only be used to specify the type of a literal constant.
Another restriction on
<TT
CLASS="LITERAL"
><VAR
CLASS="REPLACEABLE"
>type</VAR
> '<VAR
CLASS="REPLACEABLE"
>string</VAR
>'</TT
>
is that it does not work for array types; use <TT
CLASS="LITERAL"
>::</TT
>
or <TT
CLASS="LITERAL"
>CAST()</TT
> to specify the type of an array constant.
</P
></DIV
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="SQL-SYNTAX-OPERATORS"
>4.1.3. Operators</A
></H2
><A
NAME="AEN1245"
></A
><P
> An operator name is a sequence of up to <TT
CLASS="SYMBOL"
>NAMEDATALEN</TT
>-1
(63 by default) characters from the following list:
<P
CLASS="LITERALLAYOUT"
>+ - * / < > = ~ ! @ # % ^ & | ` ?</P
>
There are a few restrictions on operator names, however:
<P
></P
></P><UL
><LI
><P
> <TT
CLASS="LITERAL"
>--</TT
> and <TT
CLASS="LITERAL"
>/*</TT
> cannot appear
anywhere in an operator name, since they will be taken as the
start of a comment.
</P
></LI
><LI
><P
> A multiple-character operator name cannot end in <TT
CLASS="LITERAL"
>+</TT
> or <TT
CLASS="LITERAL"
>-</TT
>,
unless the name also contains at least one of these characters:
<P
CLASS="LITERALLAYOUT"
>~ ! @ # % ^ & | ` ?</P
>
For example, <TT
CLASS="LITERAL"
>@-</TT
> is an allowed operator name,
but <TT
CLASS="LITERAL"
>*-</TT
> is not. This restriction allows
<SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> to parse SQL-compliant
queries without requiring spaces between tokens.
</P
></LI
></UL
><P>
</P
><P
> When working with non-SQL-standard operator names, you will usually
need to separate adjacent operators with spaces to avoid ambiguity.
For example, if you have defined a left unary operator named <TT
CLASS="LITERAL"
>@</TT
>,
you cannot write <TT
CLASS="LITERAL"
>X*@Y</TT
>; you must write
<TT
CLASS="LITERAL"
>X* @Y</TT
> to ensure that
<SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
> reads it as two operator names
not one.
</P
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="AEN1269"
>4.1.4. Special Characters</A
></H2
><P
> Some characters that are not alphanumeric have a special meaning
that is different from being an operator. Details on the usage can
be found at the location where the respective syntax element is
described. This section only exists to advise the existence and
summarize the purposes of these characters.
<P
></P
></P><UL
><LI
><P
> A dollar sign (<TT
CLASS="LITERAL"
>$</TT
>) followed by digits is used
to represent a positional parameter in the body of a function
definition or a prepared statement. In other contexts the
dollar sign may be part of an identifier.
</P
></LI
><LI
><P
> Parentheses (<TT
CLASS="LITERAL"
>()</TT
>) have their usual meaning to
group expressions and enforce precedence. In some cases
parentheses are required as part of the fixed syntax of a
particular SQL command.
</P
></LI
><LI
><P
> Brackets (<TT
CLASS="LITERAL"
>[]</TT
>) are used to select the elements
of an array. See <A
HREF="arrays.html"
>Section 8.10</A
> for more information
on arrays.
</P
></LI
><LI
><P
> Commas (<TT
CLASS="LITERAL"
>,</TT
>) are used in some syntactical
constructs to separate the elements of a list.
</P
></LI
><LI
><P
> The semicolon (<TT
CLASS="LITERAL"
>;</TT
>) terminates an SQL command.
It cannot appear anywhere within a command, except within a
string constant or quoted identifier.
</P
></LI
><LI
><P
> The colon (<TT
CLASS="LITERAL"
>:</TT
>) is used to select
<SPAN
CLASS="QUOTE"
>"slices"</SPAN
> from arrays. (See <A
HREF="arrays.html"
>Section 8.10</A
>.) In certain SQL dialects (such as Embedded
SQL), the colon is used to prefix variable names.
</P
></LI
><LI
><P
> The asterisk (<TT
CLASS="LITERAL"
>*</TT
>) has a special meaning when
used in the <TT
CLASS="COMMAND"
>SELECT</TT
> command or with the
<CODE
CLASS="FUNCTION"
>COUNT</CODE
> aggregate function.
</P
></LI
><LI
><P
> The period (<TT
CLASS="LITERAL"
>.</TT
>) is used in numeric
constants, and to separate schema, table, and column names.
</P
></LI
></UL
><P>
</P
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="SQL-SYNTAX-COMMENTS"
>4.1.5. Comments</A
></H2
><A
NAME="AEN1304"
></A
><P
> A comment is an arbitrary sequence of characters beginning with
double dashes and extending to the end of the line, e.g.:
</P><PRE
CLASS="PROGRAMLISTING"
>-- This is a standard SQL comment</PRE
><P>
</P
><P
> Alternatively, C-style block comments can be used:
</P><PRE
CLASS="PROGRAMLISTING"
>/* multiline comment
* with nesting: /* nested block comment */
*/</PRE
><P>
where the comment begins with <TT
CLASS="LITERAL"
>/*</TT
> and extends to
the matching occurrence of <TT
CLASS="LITERAL"
>*/</TT
>. These block
comments nest, as specified in the SQL standard but unlike C, so that one can
comment out larger blocks of code that may contain existing block
comments.
</P
><P
> A comment is removed from the input stream before further syntax
analysis and is effectively replaced by whitespace.
</P
></DIV
><DIV
CLASS="SECT2"
><H2
CLASS="SECT2"
><A
NAME="SQL-PRECEDENCE"
>4.1.6. Lexical Precedence</A
></H2
><A
NAME="AEN1316"
></A
><P
> <A
HREF="sql-syntax.html#SQL-PRECEDENCE-TABLE"
>Table 4-1</A
> shows the precedence and
associativity of the operators in <SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
>.
Most operators have the same precedence and are left-associative.
The precedence and associativity of the operators is hard-wired
into the parser. This may lead to non-intuitive behavior; for
example the Boolean operators <TT
CLASS="LITERAL"
><</TT
> and
<TT
CLASS="LITERAL"
>></TT
> have a different precedence than the Boolean
operators <TT
CLASS="LITERAL"
><=</TT
> and <TT
CLASS="LITERAL"
>>=</TT
>. Also, you will
sometimes need to add parentheses when using combinations of
binary and unary operators. For instance
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT 5 ! - 6;</PRE
><P>
will be parsed as
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT 5 ! (- 6);</PRE
><P>
because the parser has no idea -- until it is too late -- that
<TT
CLASS="TOKEN"
>!</TT
> is defined as a postfix operator, not an infix one.
To get the desired behavior in this case, you must write
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT (5 !) - 6;</PRE
><P>
This is the price one pays for extensibility.
</P
><DIV
CLASS="TABLE"
><A
NAME="SQL-PRECEDENCE-TABLE"
></A
><P
><B
>Table 4-1. Operator Precedence (decreasing)</B
></P
><TABLE
BORDER="1"
CLASS="CALSTABLE"
><COL><COL><COL><THEAD
><TR
><TH
>Operator/Element</TH
><TH
>Associativity</TH
><TH
>Description</TH
></TR
></THEAD
><TBODY
><TR
><TD
><TT
CLASS="TOKEN"
>.</TT
></TD
><TD
>left</TD
><TD
>table/column name separator</TD
></TR
><TR
><TD
><TT
CLASS="TOKEN"
>::</TT
></TD
><TD
>left</TD
><TD
><SPAN
CLASS="PRODUCTNAME"
>PostgreSQL</SPAN
>-style typecast</TD
></TR
><TR
><TD
><TT
CLASS="TOKEN"
>[</TT
> <TT
CLASS="TOKEN"
>]</TT
></TD
><TD
>left</TD
><TD
>array element selection</TD
></TR
><TR
><TD
><TT
CLASS="TOKEN"
>-</TT
></TD
><TD
>right</TD
><TD
>unary minus</TD
></TR
><TR
><TD
><TT
CLASS="TOKEN"
>^</TT
></TD
><TD
>left</TD
><TD
>exponentiation</TD
></TR
><TR
><TD
><TT
CLASS="TOKEN"
>*</TT
> <TT
CLASS="TOKEN"
>/</TT
> <TT
CLASS="TOKEN"
>%</TT
></TD
><TD
>left</TD
><TD
>multiplication, division, modulo</TD
></TR
><TR
><TD
><TT
CLASS="TOKEN"
>+</TT
> <TT
CLASS="TOKEN"
>-</TT
></TD
><TD
>left</TD
><TD
>addition, subtraction</TD
></TR
><TR
><TD
><TT
CLASS="TOKEN"
>IS</TT
></TD
><TD
> </TD
><TD
><TT
CLASS="LITERAL"
>IS TRUE</TT
>, <TT
CLASS="LITERAL"
>IS FALSE</TT
>, <TT
CLASS="LITERAL"
>IS UNKNOWN</TT
>, <TT
CLASS="LITERAL"
>IS NULL</TT
></TD
></TR
><TR
><TD
><TT
CLASS="TOKEN"
>ISNULL</TT
></TD
><TD
> </TD
><TD
>test for null</TD
></TR
><TR
><TD
><TT
CLASS="TOKEN"
>NOTNULL</TT
></TD
><TD
> </TD
><TD
>test for not null</TD
></TR
><TR
><TD
>(any other)</TD
><TD
>left</TD
><TD
>all other native and user-defined operators</TD
></TR
><TR
><TD
><TT
CLASS="TOKEN"
>IN</TT
></TD
><TD
> </TD
><TD
>set membership</TD
></TR
><TR
><TD
><TT
CLASS="TOKEN"
>BETWEEN</TT
></TD
><TD
> </TD
><TD
>containment</TD
></TR
><TR
><TD
><TT
CLASS="TOKEN"
>OVERLAPS</TT
></TD
><TD
> </TD
><TD
>time interval overlap</TD
></TR
><TR
><TD
><TT
CLASS="TOKEN"
>LIKE</TT
> <TT
CLASS="TOKEN"
>ILIKE</TT
> <TT
CLASS="TOKEN"
>SIMILAR</TT
></TD
><TD
> </TD
><TD
>string pattern matching</TD
></TR
><TR
><TD
><TT
CLASS="TOKEN"
><</TT
> <TT
CLASS="TOKEN"
>></TT
></TD
><TD
> </TD
><TD
>less than, greater than</TD
></TR
><TR
><TD
><TT
CLASS="TOKEN"
>=</TT
></TD
><TD
>right</TD
><TD
>equality, assignment</TD
></TR
><TR
><TD
><TT
CLASS="TOKEN"
>NOT</TT
></TD
><TD
>right</TD
><TD
>logical negation</TD
></TR
><TR
><TD
><TT
CLASS="TOKEN"
>AND</TT
></TD
><TD
>left</TD
><TD
>logical conjunction</TD
></TR
><TR
><TD
><TT
CLASS="TOKEN"
>OR</TT
></TD
><TD
>left</TD
><TD
>logical disjunction</TD
></TR
></TBODY
></TABLE
></DIV
><P
> Note that the operator precedence rules also apply to user-defined
operators that have the same names as the built-in operators
mentioned above. For example, if you define a
<SPAN
CLASS="QUOTE"
>"+"</SPAN
> operator for some custom data type it will have
the same precedence as the built-in <SPAN
CLASS="QUOTE"
>"+"</SPAN
> operator, no
matter what yours does.
</P
><P
> When a schema-qualified operator name is used in the
<TT
CLASS="LITERAL"
>OPERATOR</TT
> syntax, as for example in
</P><PRE
CLASS="PROGRAMLISTING"
>SELECT 3 OPERATOR(pg_catalog.+) 4;</PRE
><P>
the <TT
CLASS="LITERAL"
>OPERATOR</TT
> construct is taken to have the default precedence
shown in <A
HREF="sql-syntax.html#SQL-PRECEDENCE-TABLE"
>Table 4-1</A
> for <SPAN
CLASS="QUOTE"
>"any other"</SPAN
> operator. This is true no matter
which specific operator name appears inside <TT
CLASS="LITERAL"
>OPERATOR()</TT
>.
</P
></DIV
></DIV
></DIV
><DIV
CLASS="NAVFOOTER"
><HR
ALIGN="LEFT"
WIDTH="100%"><TABLE
SUMMARY="Footer navigation table"
WIDTH="100%"
BORDER="0"
CELLPADDING="0"
CELLSPACING="0"
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
><A
HREF="sql.html"
ACCESSKEY="P"
>Prev</A
></TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
><A
HREF="index.html"
ACCESSKEY="H"
>Home</A
></TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
><A
HREF="sql-expressions.html"
ACCESSKEY="N"
>Next</A
></TD
></TR
><TR
><TD
WIDTH="33%"
ALIGN="left"
VALIGN="top"
>The SQL Language</TD
><TD
WIDTH="34%"
ALIGN="center"
VALIGN="top"
><A
HREF="sql.html"
ACCESSKEY="U"
>Up</A
></TD
><TD
WIDTH="33%"
ALIGN="right"
VALIGN="top"
>Value Expressions</TD
></TR
></TABLE
></DIV
></BODY
></HTML
>
