Skip to content

P
Postgres FD Implementation
Commit bf886d5b authored by Peter Eisentraut's avatar Peter Eisentraut
Browse files
Options

Add section explaining unspecified expression evaluation order.

parent 1731048c
Hide whitespace changes
Inline Side-by-side
Showing with 210 additions and 183 deletions
doc/src/sgml/syntax.sgml
View file @ bf886d5b
<!-- <!--
$Header: /cvsroot/pgsql/doc/src/sgml/syntax.sgml,v 1.60 2002/04/25 20:14:43 tgl Exp $ $Header: /cvsroot/pgsql/doc/src/sgml/syntax.sgml,v 1.61 2002/06/01 20:56:55 petere Exp $
--> -->
<chapter id="sql-syntax"> <chapter id="sql-syntax">
...@@ -44,7 +44,6 @@ $Header: /cvsroot/pgsql/doc/src/sgml/syntax.sgml,v 1.60 2002/04/25 20:14:43 tgl ...@@ -44,7 +44,6 @@ $Header: /cvsroot/pgsql/doc/src/sgml/syntax.sgml,v 1.60 2002/04/25 20:14:43 tgl
whitespace. whitespace.
</para> </para>
<informalexample id="sql-syntax-ex-commands">
<para> <para>
For example, the following is (syntactically) valid SQL input: For example, the following is (syntactically) valid SQL input:
<programlisting> <programlisting>
...@@ -56,7 +55,6 @@ INSERT INTO MY_TABLE VALUES (3, 'hi there'); ...@@ -56,7 +55,6 @@ INSERT INTO MY_TABLE VALUES (3, 'hi there');
is not required; more than one command can be on a line, and is not required; more than one command can be on a line, and
commands can usefully be split across lines). commands can usefully be split across lines).
</para> </para>
</informalexample>
<para> <para>
The SQL syntax is not very consistent regarding what tokens The SQL syntax is not very consistent regarding what tokens
...@@ -328,18 +326,16 @@ SELECT 'foo' 'bar'; ...@@ -328,18 +326,16 @@ SELECT 'foo' 'bar';
characters embedded in the constant. characters embedded in the constant.
</para> </para>
<informalexample> <para>
<para> These are some examples of valid floating-point constants:
These are some examples of valid floating-point constants: <literallayout>
<literallayout>
3.5 3.5
4. 4.
.001 .001
5e2 5e2
1.925e-3 1.925e-3
</literallayout> </literallayout>
</para> </para>
</informalexample>
<para> <para>
Floating-point constants are of type <type>DOUBLE Floating-point constants are of type <type>DOUBLE
...@@ -347,10 +343,10 @@ SELECT 'foo' 'bar'; ...@@ -347,10 +343,10 @@ SELECT 'foo' 'bar';
by using <acronym>SQL</acronym> string notation or by using <acronym>SQL</acronym> string notation or
<productname>PostgreSQL</productname> type notation: <productname>PostgreSQL</productname> type notation:
<programlisting> <programlisting>
REAL '1.23' -- string style REAL '1.23' -- string style
'1.23'::REAL -- PostgreSQL (historical) style '1.23'::REAL -- PostgreSQL (historical) style
</programlisting> </programlisting>
</para> </para>
</sect3> </sect3>
...@@ -621,6 +617,184 @@ CAST ( '<replaceable>string</replaceable>' AS <replaceable>type</replaceable> ) ...@@ -621,6 +617,184 @@ CAST ( '<replaceable>string</replaceable>' AS <replaceable>type</replaceable> )
analysis and is effectively replaced by whitespace. analysis and is effectively replaced by whitespace.
</para> </para>
</sect2> </sect2>
<sect2 id="sql-precedence">
<title>Lexical Precedence</title>
<indexterm zone="sql-precedence">
<primary>operators</primary>
<secondary>precedence</secondary>
</indexterm>
<para>
The precedence and associativity of the operators is hard-wired
into the parser. Most operators have the same precedence and are
left-associative. This may lead to non-intuitive behavior; for
example the Boolean operators <literal>&lt;</> and <literal>&gt;</> have a different
precedence than the Boolean operators <literal>&lt;=</> and <literal>&gt;=</>. Also,
you will sometimes need to add parentheses when using combinations
of binary and unary operators. For instance
<programlisting>
SELECT 5 ! - 6;
</programlisting>
will be parsed as
<programlisting>
SELECT 5 ! (- 6);
</programlisting>
because the parser has no idea -- until it is too late -- that
<token>!</token> is defined as a postfix operator, not an infix one.
To get the desired behavior in this case, you must write
<programlisting>
SELECT (5 !) - 6;
</programlisting>
This is the price one pays for extensibility.
</para>
<table tocentry="1">
<title>Operator Precedence (decreasing)</title>
<tgroup cols="3">
<thead>
<row>
<entry>Operator/Element</entry>
<entry>Associativity</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry><token>.</token></entry>
<entry>left</entry>
<entry>table/column name separator</entry>
</row>
<row>
<entry><token>::</token></entry>
<entry>left</entry>
<entry><productname>PostgreSQL</productname>-style typecast</entry>
</row>
<row>
<entry><token>[</token> <token>]</token></entry>
<entry>left</entry>
<entry>array element selection</entry>
</row>
<row>
<entry><token>-</token></entry>
<entry>right</entry>
<entry>unary minus</entry>
</row>
<row>
<entry><token>^</token></entry>
<entry>left</entry>
<entry>exponentiation</entry>
</row>
<row>
<entry><token>*</token> <token>/</token> <token>%</token></entry>
<entry>left</entry>
<entry>multiplication, division, modulo</entry>
</row>
<row>
<entry><token>+</token> <token>-</token></entry>
<entry>left</entry>
<entry>addition, subtraction</entry>
</row>
<row>
<entry><token>IS</token></entry>
<entry></entry>
<entry>test for TRUE, FALSE, UNKNOWN, NULL</entry>
</row>
<row>
<entry><token>ISNULL</token></entry>
<entry></entry>
<entry>test for NULL</entry>
</row>
<row>
<entry><token>NOTNULL</token></entry>
<entry></entry>
<entry>test for NOT NULL</entry>
</row>
<row>
<entry>(any other)</entry>
<entry>left</entry>
<entry>all other native and user-defined operators</entry>
</row>
<row>
<entry><token>IN</token></entry>
<entry></entry>
<entry>set membership</entry>
</row>
<row>
<entry><token>BETWEEN</token></entry>
<entry></entry>
<entry>containment</entry>
</row>
<row>
<entry><token>OVERLAPS</token></entry>
<entry></entry>
<entry>time interval overlap</entry>
</row>
<row>
<entry><token>LIKE</token> <token>ILIKE</token></entry>
<entry></entry>
<entry>string pattern matching</entry>
</row>
<row>
<entry><token>&lt;</token> <token>&gt;</token></entry>
<entry></entry>
<entry>less than, greater than</entry>
</row>
<row>
<entry><token>=</token></entry>
<entry>right</entry>
<entry>equality, assignment</entry>
</row>
<row>
<entry><token>NOT</token></entry>
<entry>right</entry>
<entry>logical negation</entry>
</row>
<row>
<entry><token>AND</token></entry>
<entry>left</entry>
<entry>logical conjunction</entry>
</row>
<row>
<entry><token>OR</token></entry>
<entry>left</entry>
<entry>logical disjunction</entry>
</row>
</tbody>
</tgroup>
</table>
<para>
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
<quote>+</quote> operator for some custom data type it will have
the same precedence as the built-in <quote>+</quote> operator, no
matter what yours does.
</para>
</sect2>
</sect1> </sect1>
<sect1 id="sql-naming"> <sect1 id="sql-naming">
...@@ -728,11 +902,11 @@ CAST ( '<replaceable>string</replaceable>' AS <replaceable>type</replaceable> ) ...@@ -728,11 +902,11 @@ CAST ( '<replaceable>string</replaceable>' AS <replaceable>type</replaceable> )
need to write a qualified operator name in an expression, there is a need to write a qualified operator name in an expression, there is a
special provision: you must write special provision: you must write
<synopsis> <synopsis>
<literal>OPERATOR(</><replaceable>schema</><literal>.</><replaceable>operator</><literal>)</> <literal>OPERATOR(</><replaceable>schema</><literal>.</><replaceable>operator</><literal>)</>
</synopsis> </synopsis>
This is needed to avoid syntactic ambiguity. An example is This is needed to avoid syntactic ambiguity. An example is
<programlisting> <programlisting>
SELECT 3 OPERATOR(pg_catalog.+) 4; SELECT 3 OPERATOR(pg_catalog.+) 4;
</programlisting> </programlisting>
In practice one usually relies on the search path for operators, In practice one usually relies on the search path for operators,
so as not to have to write anything so ugly as that. so as not to have to write anything so ugly as that.
...@@ -1256,186 +1430,39 @@ FROM states; ...@@ -1256,186 +1430,39 @@ FROM states;
</para> </para>
</sect2> </sect2>
</sect1> <sect2>
<title>Expression Evaluation</title>
<sect1 id="sql-precedence">
<title>Lexical Precedence</title>
<indexterm zone="sql-precedence"> <para>
<primary>operators</primary> The order of evaluation of subexpressions is not defined. In
<secondary>precedence</secondary> particular, subexpressions are not necessarily evaluated
</indexterm> left-to-right, right-to-left, or according to the lexical
precedence rules.
</para>
<para> <para>
The precedence and associativity of the operators is hard-wired Furthermore, if the result of an expression can be determined by
into the parser. Most operators have the same precedence and are evaluating only some parts of it, then some subexpressions
left-associative. This may lead to non-intuitive behavior; for might not be evaluated at all. For instance, if one wrote
example the Boolean operators <literal>&lt;</> and <literal>&gt;</> have a different
precedence than the Boolean operators <literal>&lt;=</> and <literal>&gt;=</>. Also,
you will sometimes need to add parentheses when using combinations
of binary and unary operators. For instance
<programlisting> <programlisting>
SELECT 5 ! - 6; SELECT true OR somefunc();
</programlisting> </programlisting>
will be parsed as then <literal>somefunc()</literal> would (probably) not be called
at all. The same would be the case if one wrote
<programlisting> <programlisting>
SELECT 5 ! (- 6); SELECT somefunc() OR true;
</programlisting> </programlisting>
because the parser has no idea -- until it is too late -- that Note that this is not the same as the left-to-right
<token>!</token> is defined as a postfix operator, not an infix one. <quote>short-circuiting</quote> of Boolean operators that is found
To get the desired behavior in this case, you must write in some programming languages.
<programlisting>
SELECT (5 !) - 6;
</programlisting>
This is the price one pays for extensibility.
</para> </para>
<table tocentry="1">
<title>Operator Precedence (decreasing)</title>
<tgroup cols="3">
<thead>
<row>
<entry>Operator/Element</entry>
<entry>Associativity</entry>
<entry>Description</entry>
</row>
</thead>
<tbody>
<row>
<entry><token>.</token></entry>
<entry>left</entry>
<entry>table/column name separator</entry>
</row>
<row>
<entry><token>::</token></entry>
<entry>left</entry>
<entry><productname>PostgreSQL</productname>-style typecast</entry>
</row>
<row>
<entry><token>[</token> <token>]</token></entry>
<entry>left</entry>
<entry>array element selection</entry>
</row>
<row>
<entry><token>-</token></entry>
<entry>right</entry>
<entry>unary minus</entry>
</row>
<row>
<entry><token>^</token></entry>
<entry>left</entry>
<entry>exponentiation</entry>
</row>
<row>
<entry><token>*</token> <token>/</token> <token>%</token></entry>
<entry>left</entry>
<entry>multiplication, division, modulo</entry>
</row>
<row>
<entry><token>+</token> <token>-</token></entry>
<entry>left</entry>
<entry>addition, subtraction</entry>
</row>
<row>
<entry><token>IS</token></entry>
<entry></entry>
<entry>test for TRUE, FALSE, UNKNOWN, NULL</entry>
</row>
<row>
<entry><token>ISNULL</token></entry>
<entry></entry>
<entry>test for NULL</entry>
</row>
<row>
<entry><token>NOTNULL</token></entry>
<entry></entry>
<entry>test for NOT NULL</entry>
</row>
<row>
<entry>(any other)</entry>
<entry>left</entry>
<entry>all other native and user-defined operators</entry>
</row>
<row>
<entry><token>IN</token></entry>
<entry></entry>
<entry>set membership</entry>
</row>
<row>
<entry><token>BETWEEN</token></entry>
<entry></entry>
<entry>containment</entry>
</row>
<row>
<entry><token>OVERLAPS</token></entry>
<entry></entry>
<entry>time interval overlap</entry>
</row>
<row>
<entry><token>LIKE</token> <token>ILIKE</token></entry>
<entry></entry>
<entry>string pattern matching</entry>
</row>
<row>
<entry><token>&lt;</token> <token>&gt;</token></entry>
<entry></entry>
<entry>less than, greater than</entry>
</row>
<row>
<entry><token>=</token></entry>
<entry>right</entry>
<entry>equality, assignment</entry>
</row>
<row>
<entry><token>NOT</token></entry>
<entry>right</entry>
<entry>logical negation</entry>
</row>
<row>
<entry><token>AND</token></entry>
<entry>left</entry>
<entry>logical conjunction</entry>
</row>
<row>
<entry><token>OR</token></entry>
<entry>left</entry>
<entry>logical disjunction</entry>
</row>
</tbody>
</tgroup>
</table>
<para> <para>
Note that the operator precedence rules also apply to user-defined As a consequence, it is unwise to use functions with side effects
operators that have the same names as the built-in operators as part of complex expressions.
mentioned above. For example, if you define a
<quote>+</quote> operator for some custom data type it will have
the same precedence as the built-in <quote>+</quote> operator, no
matter what yours does.
</para> </para>
</sect1> </sect2>
</sect1>
</chapter> </chapter>
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment