boolean.xml

<!-- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% -->
<!-- %% -->
<!-- %A  boolean.msk                 GAP documentation            Martin Schönert -->
<!-- %A                                                           Alexander Hulpke -->
<!-- %% -->
<!-- %A  @(#)<M>Id: boolean.msk,v 1.14 2002/04/15 10:02:27 sal Exp </M> -->
<!-- %% -->
<!-- %Y  (C) 1998 School Math and Comp. Sci., University of St Andrews, Scotland -->
<!-- %Y  Copyright (C) 2002 The GAP Group -->
<!-- %% -->
<Chapter Label="Booleans">
<Heading>Booleans</Heading>

<Index Subkey="boolean">type</Index>
<Index>logical</Index>

The two main <E>boolean</E> values are <K>true</K> and <K>false</K>.
They stand for the <E>logical</E> values of the same name.
They appear as values of the conditions in <K>if</K>-statements
and <K>while</K>-loops.
Booleans are also important as return values of <E>filters</E>
(see&nbsp;<Ref Sect="Filters"/>)
such as <Ref Prop="IsFinite"/> and <Ref Func="IsBool"/>.
Note that it is a convention that the name of a function that
returns <K>true</K> or <K>false</K> according to the outcome,
starts with <C>Is</C>.
<P/>
For technical reasons, also the value <K>fail</K>
(see&nbsp;<Ref Sect="Fail"/>) is regarded as a boolean.


<!-- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% -->
<Section Label="sect:IsBool">
<Heading>IsBool (Filter)</Heading>

<#Include Label="IsBool">

</Section>


<!-- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% -->
<Section Label="Fail">
<Heading>Fail (Variable)</Heading>

<ManSection>
<Var Name="fail"/>

<Description>
The value <K>fail</K> is used to indicate situations when an operation could
not be performed for the given arguments, either because of shortcomings of
the arguments or because of restrictions in the implementation or
computability.
So for example <Ref Func="Position"/> will return <K>fail</K>
if the point searched for is not in the list.
<P/>
<K>fail</K> is simply an object that is different from every other object
than itself.
<P/>
For technical reasons, <K>fail</K> is a boolean value.
But note that <K>fail</K> cannot be used to form boolean expressions with
<K>and</K>, <K>or</K>, and <K>not</K>
(see&nbsp;<Ref Sect="Operations for Booleans"/> below),
and <K>fail</K> cannot appear in boolean lists
(see Chapter&nbsp;<Ref Chap="Boolean Lists"/>).
</Description>
</ManSection>

</Section>


<!-- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% -->
<Section Label="Comparisons of Booleans">
<Heading>Comparisons of Booleans</Heading>
<Index Subkey="of booleans">comparisons</Index>
<Subsection Label="Equality and inequality of Booleans">
<Heading>Equality and inequality of Booleans</Heading>
<Index Subkey="of booleans">equality</Index>
<Index Subkey="of booleans">inequality</Index>
<C><A>bool1</A> = <A>bool2</A></C>
<P/>
<Alt Only="LaTeX">\noindent</Alt>
<C><A>bool1</A> &lt;> <A>bool2</A></C>
<P/>
The equality operator <C>=</C> evaluates to <K>true</K>
if the two boolean values <A>bool1</A> and <A>bool2</A> are equal,
i.e., both are <K>true</K> or both are <K>false</K> or both <K>fail</K>,
and <K>false</K> otherwise.
The inequality operator <C>&lt;></C> evaluates to <K>true</K>
if the two boolean values <A>bool1</A>, <A>bool2</A> are different,
and <K>false</K> otherwise.
This operation is also called the <E>exclusive or</E>,
because its value is <K>true</K> if exactly one of <A>bool1</A> or
<A>bool2</A> is <K>true</K>.
<P/>
You can compare boolean values with objects of other types.
Of course they are never equal.
<P/>
<Example><![CDATA[
gap> true = false;
false
gap> false = (true = fail);
true
gap> true <> 17;
true
]]></Example>
</Subsection>

<Subsection Label="Ordering of Booleans">
<Heading>Ordering of Booleans</Heading>
<Index Subkey="booleans">ordering</Index>
<A>bool1</A> <C>&lt;</C> <A>bool2</A>
<P/>
The ordering of boolean values is defined by
<K>true</K> <C>&lt;</C> <K>false</K> <C>&lt;</C> <K>fail</K>.
For the comparison of booleans with other &GAP; objects,
see Section&nbsp;<Ref Sect="Comparisons"/>.
<P/>
<Example><![CDATA[
gap> true < false;  fail >= false;
true
true
]]></Example>
</Subsection>

</Section>


<!-- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% -->
<Section Label="Operations for Booleans">
<Heading>Operations for Booleans</Heading>

<Index Subkey="for booleans">operations</Index>
<Index>logical operations</Index>

The following boolean operations are only applicable to <K>true</K> and
<K>false</K>.
<P/>

<Subsection Label="Logical disjunction">
<Heading>Logical disjunction</Heading>
<Index>Logical disjunction</Index>
<Index Key="or"><K>or</K></Index>
<A>bool1</A> <K>or</K> <A>bool2</A>
<P/>
The logical operator <K>or</K> evaluates to <K>true</K>
if at least one of the two boolean operands <A>bool1</A> and <A>bool2</A>
is <K>true</K>, and to <K>false</K> otherwise.
<P/>
<K>or</K> first evaluates <A>bool1</A>.
If the value is neither <K>true</K> nor <K>false</K> an error is signalled.
If the value is <K>true</K>, then <K>or</K> returns <K>true</K>
<E>without</E> evaluating <A>bool2</A>.
If the value is <K>false</K>, then <K>or</K> evaluates <A>bool2</A>.
Again, if the value is neither <K>true</K> nor <K>false</K>
an error is signalled.
Otherwise <K>or</K> returns the value of <A>bool2</A>.
This <E>short-circuited</E> evaluation is important if the value of
<A>bool1</A> is <K>true</K>
and evaluation of <A>bool2</A> would take much time or cause an error.
<P/>
<K>or</K> is associative, i.e., it is allowed to write
<A>b1</A> <K>or</K> <A>b2</A> <K>or</K> <A>b3</A>,
which is interpreted as (<A>b1</A> <K>or</K> <A>b2</A>) <K>or</K> <A>b3</A>.
<K>or</K> has the lowest precedence of the logical operators.
All logical operators have lower precedence than the comparison operators
<C>=</C>, <C>&lt;</C>, <K>in</K>, etc.
<P/>
<Example><![CDATA[
gap> true or false;
true
gap> false or false;
false
gap> i := -1;; l := [1,2,3];;
gap> if i <= 0 or l[i] = false then     # this does not cause an error,
>    Print("aha\n"); fi;                # because `l[i]' is not evaluated
aha
]]></Example>
</Subsection>

<Subsection Label="Logical conjunction">
<Heading>Logical conjunction</Heading>
<Index>Logical conjunction</Index>
<Index Key="and"><K>and</K></Index>
<A>bool1</A> <K>and</K> <A>bool2</A>
<P/>
<Index Key="and" Subkey="for filters"><K>and</K></Index>
<Alt Only="LaTeX">\noindent</Alt>
<A>fil1</A> <K>and</K> <A>fil2</A>
<P/>
The logical operator <K>and</K> evaluates to <K>true</K>
if both boolean operands <A>bool1</A>, <A>bool2</A> are <K>true</K>,
and to <K>false</K> otherwise.
<P/>
<K>and</K> first evaluates <A>bool1</A>.
If the value is neither <K>true</K> nor <K>false</K> an error is signalled.
If the value is <K>false</K>, then <K>and</K> returns <K>false</K>
<E>without</E> evaluating <A>bool2</A>.
If the value is <K>true</K>, then <K>and</K> evaluates <A>bool2</A>.
Again, if the value is neither <K>true</K> nor <K>false</K>
an error is signalled.
Otherwise <K>and</K> returns the value of <A>bool2</A>.
This <E>short-circuited</E> evaluation is important if the value of
<A>bool1</A> is <K>false</K> and evaluation of <A>bool2</A> would take much
time or cause an error.
<P/>
<K>and</K> is associative, i.e., it is allowed to write
<A>b1</A> <K>and</K> <A>b2</A> <K>and</K> <A>b3</A>,
which is interpreted as (<A>b1</A> <K>and</K> <A>b2</A>) <K>and</K> <A>b3</A>.
<K>and</K> has higher precedence than the logical <K>or</K> operator,
but lower than the unary logical <K>not</K> operator.
All logical operators have lower precedence than the comparison operators
<C>=</C>, <C>&lt;</C>, <K>in</K>, etc.
<P/>
<Example><![CDATA[
gap> true and false;
false
gap> true and true;
true
gap> false and 17;  # does not cause error, because 17 is never looked at
false
]]></Example>
<P/>
<K>and</K> can also be applied to filters.
It returns a filter that when applied to some argument <A>x</A>,
tests <A>fil1</A><M>(x)</M> <K>and</K> <A>fil2</A><M>(x)</M>.
<P/>
<Example><![CDATA[
gap> andfilt:= IsPosRat and IsInt;;
gap> andfilt( 17 );  andfilt( 1/2 );
true
false
]]></Example>
</Subsection>

<Subsection Label="Logical negation">
<Heading>Logical negation</Heading>
<Index>Logical negation</Index>
<Index Key="not"><K>not</K></Index>
<K>not</K> <A>bool</A>
<P/>
The logical operator <K>not</K> returns <K>true</K>
if the boolean value <A>bool</A> is <K>false</K>, and <K>true</K> otherwise.
An error is signalled if <A>bool</A> does not evaluate to <K>true</K> or
<K>false</K>.
<P/>
<K>not</K> has higher precedence than the other logical operators,
<K>or</K> and <K>and</K>.
All logical operators have lower precedence than the comparison operators
<C>=</C>, <C>&lt;</C>, <K>in</K>, etc.
<P/>
<Example><![CDATA[
gap> true and false;
false
gap> not true;
false
gap> not false;
true
]]></Example>
</Subsection>

</Section>
</Chapter>


<!-- %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% -->
<!-- %% -->
<!-- %E -->