Google Git
Sign in
apache / commons-jexl / e3fd7c30ae438c567b01535ce1f8e0facc75e48f / . / src / site / xdoc / reference / syntax.xml
blob: 1a8b695d66dce50fe4efe7afd18b707dc45e9453 [file] [log] [blame]
<?xml version="1.0"?>
<!--
Licensed to the Apache Software Foundation (ASF) under one or more
contributor license agreements. See the NOTICE file distributed with
this work for additional information regarding copyright ownership.
The ASF licenses this file to You under the Apache License, Version 2.0
(the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
-->
<document>
<properties>
<title>Apache Commons JEXL Syntax</title>
</properties>
<body>
<section name="Overview">
<p>
This reference is split up into the following sections:
<ol>
<li>
<a href="#Language Elements">Language Elements</a>
</li>
<li>
<a href="#Literals">Literals</a>
</li>
<li>
<a href="#Functions">Functions</a>
</li>
<li>
<a href="#Operators">Operators</a>
</li>
<li>
<a href="#Access">Access</a>
</li>
<li>
<a href="#Conditional">Conditional Statements</a>
</li>
</ol>
</p>
<p>
For more technical information about the JEXL Grammar, you can find the
<a href="https://javacc.dev.java.net/">JavaCC</a> grammar for JEXL
here: <a href="http://svn.apache.org/viewvc/commons/proper/jexl/trunk/src/main/java/org/apache/commons/jexl3/parser/Parser.jjt?view=markup">Parser.jjt</a>
</p>
</section>
<section name="Language Elements">
<table>
<tr>
<th width="15%">Item</th>
<th>Description</th>
</tr>
<tr>
<td>Comments</td>
<td>
Specified using <code>##</code> or <code>//</code>and extend to the end of line, e.g.
<code>## This is a comment</code>
Also specified using <code>//</code>, e.g.
<code>// This is a comment</code>
Multiple lines comments are specified using <code>/*...*/</code>, e.g.
<code>/* This is a
multi-line comment */</code>
</td>
</tr>
<tr>
<td>Identifiers / variables</td>
<td>
Must start with <code>a-z</code>, <code>A-Z</code>, <code>_</code> or <code>$</code>.
Can then be followed by <code>0-9</code>, <code>a-z</code>, <code>A-Z</code>, <code>_</code> or <code>$</code>.
e.g.
<ul>
<li>Valid: <code>var1</code>,<code>_a99</code>,<code>$1</code></li>
<li>Invalid: <code>9v</code>,<code>!a99</code>,<code>1$</code></li>
</ul>
<p>
Variable names are <strong>case-sensitive</strong>, e.g. <code>var1</code> and <code>Var1</code> are different variables.
</p>
<p>
<strong>NOTE:</strong> JEXL does not support variables with hyphens in them, e.g.
<code>commons-logging // invalid variable name (hyphenated)</code> is not a valid variable, but instead is treated as a
subtraction of the variable <code>logging</code> from the variable <code>commons</code>
</p>
<p>
JEXL also supports <code>ant-style</code> variables, the following is a valid variable name:
<code>my.dotted.var</code>
</p>
<p>
<strong>N.B.</strong> the following keywords are reserved, and cannot be used as a variable name or property when using the dot operator:
<code>or and eq ne lt gt le ge div mod not null true false new var break continue return</code>
For example, the following is invalid:
<code>my.new.dotted.var // invalid ('new' is keyword)</code>
In such cases, quoted identifiers or the [ ] operator can be used, for example:
<code>my.'new'.dotted.var</code>
<code>my['new'].dotted.var</code>
</p>
</td>
</tr>
<tr>
<td>Scripts</td>
<td>
<p>
A script in JEXL is made up of zero or more statements. Scripts can include one or more pragmas.
</p>
<p>
Scripts can be read from a String, File or URL.
</p>
<p>
They can be created with named parameters which allow a later evaluation to be performed with arguments.
</p>
<p>
By default a script returns the value of the last evaluated statement.
</p>
<p>
Using the <code>return</code> keyword, a script will return the expression that follows (or null).
</p>
</td>
</tr>
<tr>
<td>Local variables</td>
<td>Can be defined using the <code>var</code> keyword; their identifying rules are the same as contextual variables.
<ul>
<li>Basic declaration: <code>var x;</code></li>
<li>Declaration with assignment: <code>var theAnswer = 42;</code></li>
<li>Invalid declaration: <code>var x.y;</code></li>
</ul>
Their scope is the entire script scope and they take precedence in resolution over contextual variables.
When scripts are created with named parameters, those behave as local variables.
Local variables can not use <code>ant-style</code> naming, only one identifier.
</td>
</tr>
<tr>
<td>Statements</td>
<td>
A statement can be the empty statement, the semicolon (<code>;</code>) , block, conditional, assignment or an expression.
Statements are optionally terminated with a semicolon.
A single statement or a statement block can be annotated.
</td>
</tr>
<tr>
<td>Block</td>
<td>
A block is simply multiple statements inside curly braces (<code>{, }</code>).
</td>
</tr>
<tr>
<td>Assignment</td>
<td>
Assigns the value of a variable (<code>my.var = 'a value'</code>) using a
<code>JexlContext</code> as initial resolver. Both <em>beans</em> and <em>ant-ish</em>
variables assignment are supported.
</td>
</tr>
<tr>
<td>Expression</td>
<td>
An expression can be the literal, variable, access operator, function definition, function call, method call or
an evaluation operator.
</td>
</tr>
<tr>
<td>Function definition</td>
<td>
Defines a function within the script, usually associated with a local variable assignment.
<code>var fun = function(x, y) { x + y }</code>
The following syntax is also supported
<code>var fun = (x, y) -> { x + y }</code>
Calling a function follows the usual convention:
<code>fun(17, 25)</code>
<p>Note that functions can use local variables and parameters from their declaring script.
Those variables values are bound in the function environment at definition time.</p>
<code>var t = 20; var s = function(x, y) {x + y + t}; t = 54; s(15, 7)</code>
The function closure hoists 't' when defined; the result of the evaluation will
lead to <code>15 +7 + 20 = 42</code>.
</td>
</tr>
<tr>
<td>Method call</td>
<td>
Calls a method of an object, e.g.
<code>"hello world".hashCode()</code> will call the <code>hashCode</code> method
of the <code>"hello world"</code> String.
<p>In case of multiple arguments and overloading, JEXL will make the best effort to find
the most appropriate non ambiguous method to call.</p>
</td>
</tr>
<tr>
<td>Access Operator</td>
<td>
Allows to evaluate a property of an object, a value of the collection or an array
by using either square brackets or a dotted numeral, e.g.
<code>foo.bar</code> will access the <code>bar</code> property
of the <code>foo</code> Object.
<code>arr1[0]</code> will access the first element of the
of the <code>arr1</code> array.
<p>Access operators can be overloaded in <code>JexlArithmetic</code>, so that
the operator behaviour will differ depending on the type of the operator arguments</p>
</td>
</tr>
<tr>
<td>Evaluation Operator</td>
<td>
Performs computational, logical or comparative action between one, two or three arguments
whose values are expressions, e.g.
<code>40 + 2</code> will call the <code>add</code> operator
between two integer literals.
<p>All operators, except when stated otherwise, can be overloaded in <code>JexlArithmetic</code>, so that the action taken
will differ depending on the type of the operator arguments</p>
</td>
</tr>
<tr>
<td>#pragma</td>
<td>
Declares a pragma, a method to communicate information from a script to its execution environment, e.g.
<code>#pragma execution.option 42</code> will declare a pragma named <code>execution.option</code> with
a value of <code>42</code>.
<p>Pragma keys can be identifiers or antish names, pragma values can be literals (boolean, integer,
real, string, null, NaN) and antish names</p>
</td>
</tr>
<tr>
<td>@annotation</td>
<td>
Annotations in JEXL are 'meta-statements'; they allow to wrap the execution of the JEXL statement in a user provided
caller; typical example would be: <code>@synchronized(x) x.someMethod();</code>
<p>
Annotations may be declared with zero or more parameters;
<code>@lenient x.someMethod();</code>
<code>@synchronized(x) x.someMethod();</code>
<code>@parallel(pool, 8) x.someMethod();</code>
</p>
<p>
They also can be chained as in:
<code>@lenient @silent x.someMethod();</code>
</p>
<p>
Annotation processing is implemented by providing a JexlContext.AnnotationProcessor; its processAnnotation
method will call the annotated statement encapsulated in a Callable. Annotation arguments are evaluated
and passed as arguments to processAnnotation.
</p>
</td>
</tr>
</table>
</section>
<section name="Literals">
<table>
<tr>
<th width="15%">Item</th>
<th>Description</th>
</tr>
<tr>
<td>Integer Literals</td>
<td>1 or more digits from <code>0</code> to <code>9</code>, eg <code>42</code>.
</td>
</tr>
<tr>
<td>Float Literals</td>
<td>
1 or more digits from <code>0</code> to <code>9</code>, followed
by a decimal point and then one or more digits from
<code>0</code> to <code>9</code>,
optionally followed by <code>f</code> or <code>F</code>,
eg <code>42.0</code> or <code>42.0f</code>.
</td>
</tr>
<tr>
<td>Long Literals</td>
<td>1 or more digits from <code>0</code> to <code>9</code> suffixed with <code>l</code> or <code>L</code>
, eg <code>42l</code>.
</td>
</tr>
<tr>
<td>Double Literals</td>
<td>
1 or more digits from <code>0</code> to <code>9</code>, followed
by a decimal point and then one or more digits from
<code>0</code> to <code>9</code> suffixed with <code>d</code> or <code>D</code>
, eg <code>42.0d</code>.
</td>
</tr>
<tr>
<td>Big Integer Literals</td>
<td>1 or more digits from <code>0</code> to <code>9</code> suffixed with <code>h</code> or <code>H</code>
(for Huge ala OGNL, "does not interfere with hexa-decimal digits"), eg <code>42h</code>.
</td>
</tr>
<tr>
<td>Big Decimal Literals</td>
<td>
1 or more digits from <code>0</code> to <code>9</code>, followed
by a decimal point and then one or more digits from
<code>0</code> to <code>9</code> suffixed with <code>b</code> or <code>B</code>)
, eg <code>42.0b</code>.
</td>
</tr>
<tr>
<td>Natural literals - octal and hex support</td>
<td>
Natural numbers (i.e. Integer, Long, BigInteger) can also be expressed as octal or hexadecimal using the same format as Java.
i.e. prefix the number with <code>0</code> for octal, and prefix with <code>0x</code> or <code>0X</code> for hexadecimal.
For example <code>010</code> or <code>0x10</code>.
</td>
</tr>
<tr>
<td>Real literals - exponent support</td>
<td>
Real numbers (i.e. Float, Double, BigDecimal) can also be expressed using standard Java exponent notation.
i.e. suffix the number with <code>e</code> or <code>E</code> followed by the sign <code>+</code> or <code>-</code>
followed by one or more decimal digits.
For example <code>42.0E-1D</code> or <code>42.0E+3B</code>.
</td>
</tr>
<tr>
<td>String literals</td>
<td>
Can start and end with either <code>'</code> or <code>"</code> delimiters, e.g.
<code>"Hello world"</code> and
<code>'Hello world'</code> are equivalent.
<p>The escape character is <code>\</code> (backslash); it only escapes the string delimiter</p>
</td>
</tr>
<tr>
<td>Multiline format literals</td>
<td>
Start and end with <code>`</code> delimiter - back-quote -, e.g. <code>`Hello world`</code>
<p>The escape character is <code>\</code> (backslash); it only escapes the string delimiter.</p>
These format literals can span multiple lines and allow Unified JEXL expressions (JSTL like expressions)
to be interpolated. If a variable <code>user</code> valued <code>JEXL</code>is present in the environment - whether
as a local or global variable -, the format <code>`Hello ${user}`</code> will evaluate as <code>Hello JEXL</code>.
</td>
</tr>
<tr>
<td>Boolean literals</td>
<td>
The literals <code>true</code> and <code>false</code> can be used, e.g.
<code>val1 == true</code>
</td>
</tr>
<tr>
<td>Null literal</td>
<td>
The null value is represented as in java using the literal <code>null</code>, e.g.
<code>val1 == null</code>
</td>
</tr>
<tr>
<td>Array literal</td>
<td>
A <code>[</code> followed by one or more expressions separated by <code>,</code> and ending
with <code>]</code>, e.g.
<code>[ 1, 2, "three" ]</code>
<p>This syntax creates an <code>Object[]</code>.</p>
<p>
JEXL will attempt to strongly type the array; if all entries are of the same class or if all
entries are Number instance, the array literal will be an <code>MyClass[]</code> in the former
case, a <code>Number[]</code> in the latter case.</p>
<p>Furthermore, if all entries in the array literal are of the same class
and that class has an equivalent primitive type, the array returned will be a primitive array. e.g.
<code>[1, 2, 3]</code> will be interpreted as <code>int[]</code>.</p>
</td>
</tr>
<tr>
<td>List literal</td>
<td>
A <code>[</code> followed by one or more expressions separated by <code>,</code> and ending
with <code>,...]</code>, e.g.
<code>[ 1, 2, "three",...]</code>
<p>This syntax creates an <code>ArrayList&lt;Object&gt;</code>.</p>
</td>
</tr>
<tr>
<td>Set literal</td>
<td>
A <code>{</code> followed by one or more expressions separated by <code>,</code> and ending
with <code>}</code>, e.g.
<code>{ "one" , 2, "more"}</code>
<p>This syntax creates a <code>HashSet&lt;Object&gt;</code>.</p>
</td>
</tr>
<tr>
<td>Map literal</td>
<td>
A <code>{</code> followed by one or more sets of <code>key : value</code> pairs separated by <code>,</code> and ending
with <code>}</code>, e.g.
<code>{ "one" : 1, "two" : 2, "three" : 3, "more": "many more" }</code>
<p>This syntax creates a <code>HashMap&lt;Object,Object&gt;</code>.</p>
</td>
</tr>
<tr>
<td>Range literal</td>
<td>
A value followed by <code>..</code> and ending with other value, e.g.
<code>1 .. 42</code>
<p>This syntax creates a 'range' object in the form of a java iterable which can be used in for statement, e.g.
<code>for (i : 1..42) a = a + b[i]</code></p>
</td>
</tr>
</table>
</section>
<section name="Functions">
<table>
<tr>
<th width="15%">Function</th>
<th>Description</th>
</tr>
<tr>
<td>empty</td>
<td>
Evaluates whether an expression if 'empty'.
This is true when the argument is:
<ol>
<li>
<code>null</code>
</li>
<li>An instance of class C and the derived JexlArithmetic overloads a method 'public boolean empty(C arg)'
that returns true when the argument is considered empty</li>
<li>An empty string</li>
<li>An array of length zero</li>
<li>A collection of size zero</li>
<li>An empty map</li>
<li>Defining a method 'public boolean isEmpty()'
that returns true when the instance is considered empty</li>
</ol>
This is false in other cases (besides errors).
<code>empty(arg)</code>
</td>
</tr>
<tr>
<td>size</td>
<td>
Evaluates the 'size' of an expression.
This returns:
<ol>
<li>0 if the argument is null</li>
<li>The result of calling a method from a derived JexlArithmetic overload 'public int size(C arg)',
C being the class of the argument</li>
<li>Length of an array</li>
<li>Length of a string</li>
<li>Size of a Collection</li>
<li>Size of a Map</li>
<li>The result of calling a method 'public int size()' defined by the argument class</li>
</ol>
This returns 0 in other cases (besides errors).
<code>size("Hello")</code> returns 5.
</td>
</tr>
<tr>
<td>new</td>
<td>
Creates a new instance using a fully-qualified class name or Class:
<code>new("java.lang.Double", 10)</code> returns 10.0.
<p>Note that the first argument of <code>new</code> can be a variable or any
expression evaluating as a String or Class; the rest of the arguments are used
as arguments to the constructor for the class considered.</p>
<p>In case of multiple constructors, JEXL will make the best effort to find
the most appropriate non ambiguous constructor to call.</p>
</td>
</tr>
<tr>
<td>Top level function</td>
<td>
Top level function is a function which can be invoked without specifying a namespace.
<p>Top level function can be defined by the function definition method inside the script</p>
<p>A <code>JexlContext</code> can define methods which can be invoked as top level functions.
This can allow expressions like:
<code>string(23.0)</code></p>
<p>Another way to define top level function is to register to <code>JexlEngine</code> objects or classes
with null namespace.</p>
</td>
</tr>
<tr>
<td>ns:function</td>
<td>
A <code>JexlEngine</code> can register objects or classes used as function namespaces.
This can allow expressions like:
<code>math:cosinus(23.0)</code>
</td>
</tr>
</table>
</section>
<section name="Operators">
<table>
<tr>
<th width="15%">Operator</th>
<th>Description</th>
</tr>
<tr>
<td>Boolean <code>and</code></td>
<td>
<p>The usual <code>&amp;&amp;</code> operator can be used as well as the word <code>and</code>, e.g.
<code>cond1 and cond2</code> and
<code>cond1 &amp;&amp; cond2</code> are equivalent.</p>
<p>Note that this operator can not be overloaded</p>
</td>
</tr>
<tr>
<td>Boolean <code>or</code></td>
<td>
<p>The usual <code>||</code> operator can be used as well as the word <code>or</code>, e.g.
<code>cond1 or cond2</code> and
<code>cond1 || cond2</code> are equivalent.</p>
<p>Note that this operator can not be overloaded</p>
</td>
</tr>
<tr>
<td>Boolean <code>not</code></td>
<td>
<p>The usual <code>!</code> operator can be used as well as the word <code>not</code>, e.g.
<code>!cond1</code> and
<code>not cond1</code> are equivalent.</p>
<p>Note that this operator can not be overloaded</p>
</td>
</tr>
<tr>
<td>Bitwise <code>and</code></td>
<td>
The usual <code>&amp;</code> operator is used, e.g.
<code>33 &amp; 4</code>, 0010 0001 &amp; 0000 0100 = 0.
</td>
</tr>
<tr>
<td>Bitwise <code>or</code></td>
<td>
The usual <code>|</code> operator is used, e.g.
<code>33 | 4</code>, 0010 0001 | 0000 0100 = 0010 0101 = 37.
</td>
</tr>
<tr>
<td>Bitwise <code>xor</code></td>
<td>
The usual <code>^</code> operator is used, e.g.
<code>33 ^ 4</code>, 0010 0001 ^ 0000 0100 = 0010 0100 = 37.
</td>
</tr>
<tr>
<td>Bitwise <code>complement</code></td>
<td>
The usual <code>~</code> operator is used, e.g.
<code>~33</code>, ~0010 0001 = 1101 1110 = -34.
</td>
</tr>
<tr>
<td>Ternary conditional <code>?:</code> </td>
<td>
The usual ternary conditional operator <code>condition ? if_true : if_false</code> operator can be
used as well as the abbreviation <code>value ?: if_false</code> which returns the <code>value</code> if
its evaluation is defined, non-null and non-false, e.g.
<code>val1 ? val1 : val2</code> and
<code>val1 ?: val2 </code> are equivalent.
<p>
<strong>NOTE:</strong> The condition will evaluate to <code>false</code> when it
refers to an undefined variable or <code>null</code> for all <code>JexlEngine</code>
flag combinations. This allows explicit syntactic leniency and treats the condition
'if undefined or null or false' the same way in all cases.
</p>
<p>Note that this operator can not be overloaded</p>
</td>
</tr>
<tr>
<td>Null coalescing operator <code>??</code> </td>
<td>
The null coalescing operator returns the result of its first operand if it is defined and is not null.
<p>When <code>x</code>and<code>y</code>are null or undefined,
<code>x ?? 'unknown or null x'</code> evaluates as <code>'unknown or null x'</code>
<code>y ?? "default"</code> evaluates as <code>"default"</code>.
</p>
<p>
When <code>var x = 42</code> and <code>var y = "forty-two"</code>,<code>x??"other"</code>
evaluates as <code>42</code> and <code>y??"other"</code> evaluates as <code>"forty-two"</code>.
</p>
<p>
<strong>NOTE:</strong> this operator does not behave like the ternary conditional since it
does not coerce the first argument to a boolean to evaluate the condition.
When <code>var x = false</code> and <code>var y = 0</code>,<code>x??true</code>
evaluates as <code>false</code> and <code>y??1</code> evaluates as <code>0</code>.
</p>
<p>Note that this operator can not be overloaded</p>
</td>
</tr>
<tr>
<td>Equality</td>
<td>
The usual <code>==</code> operator can be used as well as the abbreviation <code>eq</code>.
For example
<code>val1 == val2</code> and
<code>val1 eq val2</code> are equivalent.
<ol>
<li>
<code>null</code> is only ever equal to null, that is if you compare null
to any non-null value, the result is false.
</li>
<li>Equality uses the java <code>equals</code> method</li>
</ol>
</td>
</tr>
<tr>
<td>Inequality</td>
<td>
The usual <code>!=</code> operator can be used as well as the abbreviation <code>ne</code>.
For example
<code>val1 != val2</code> and
<code>val1 ne val2</code> are equivalent.
</td>
</tr>
<tr>
<td>Less Than</td>
<td>
The usual <code>&lt;</code> operator can be used as well as the abbreviation <code>lt</code>.
For example
<code>val1 &lt; val2</code> and
<code>val1 lt val2</code> are equivalent.
</td>
</tr>
<tr>
<td>Less Than Or Equal To</td>
<td>
The usual <code>&lt;=</code> operator can be used as well as the abbreviation <code>le</code>.
For example
<code>val1 &lt;= val2</code> and
<code>val1 le val2</code> are equivalent.
</td>
</tr>
<tr>
<td>Greater Than</td>
<td>
The usual <code>&gt;</code> operator can be used as well as the abbreviation <code>gt</code>.
For example
<code>val1 &gt; val2</code> and
<code>val1 gt val2</code> are equivalent.
</td>
</tr>
<tr>
<td>Greater Than Or Equal To</td>
<td>
The usual <code>&gt;=</code> operator can be used as well as the abbreviation <code>ge</code>.
For example
<code>val1 &gt;= val2</code> and
<code>val1 ge val2</code> are equivalent.
</td>
</tr>
<tr>
<td>In or Match<code>=~</code></td>
<td>
The syntactically Perl inspired <code>=~</code> operator can be used to check that a <code>string</code> matches
a regular expression (expressed either a Java String or a java.util.regex.Pattern).
For example
<code>"abcdef" =~ "abc.*</code> returns <code>true</code>.
It also checks whether any collection, set or map (on keys) contains a value or not; in that case, it behaves
as an "in" operator.
Note that arrays and user classes exposing a public 'contains' method will allow their instances
to behave as right-hand side operands of this operator.
<code> "a" =~ ["a","b","c","d","e",f"]</code> returns <code>true</code>.
</td>
</tr>
<tr>
<td>Not-In or Not-Match<code>!~</code></td>
<td>
The syntactically Perl inspired <code>!~</code> operator can be used to check that a <code>string</code> does not
match a regular expression (expressed either a Java String or a java.util.regex.Pattern).
For example
<code>"abcdef" !~ "abc.*</code> returns <code>false</code>.
It also checks whether any collection, set or map (on keys) does not contain a value; in that case, it behaves
as "not in" operator.
Note that arrays and user classes exposing a public 'contains' method will allow their instances
to behave as right-hand side operands of this operator.
<code> "a" !~ ["a","b","c","d","e",f"]</code> returns <code>true</code>.
</td>
</tr>
<tr>
<td>Starts With<code>=^</code></td>
<td>
The <code>=^</code> operator is a short-hand for the 'startsWith' method.
For example, <code> "abcdef" =^ "abc" </code> returns <code>true</code>.
Note that through duck-typing, user classes exposing a public 'startsWith' method will allow their instances
to behave as left-hand side operands of this operator.
</td>
</tr>
<tr>
<td>Not Starts With<code>!^</code></td>
<td>
This is the negation of the 'starts with' operator.
<code>a !^ "abc"</code> is equivalent to <code>!(a =^ "abc")</code>
</td>
</tr>
<tr>
<td>Ends With<code>=$</code></td>
<td>The <code>=$</code> operator is a short-hand for the 'endsWith' method.
For example, <code> "abcdef" =$ "def" </code> returns <code>true</code>.
Note that through duck-typing, user classes exposing an 'endsWith' method will allow their instances
to behave as left-hand side operands of this operator.
</td>
</tr>
<tr>
<td>Not Ends With<code>!$</code></td>
<td>
This is the negation of the 'ends with' operator.
<code>a !$ "abc"</code> is equivalent to <code>!(a =$ "abc")</code>
</td>
</tr>
<tr>
<td>Addition</td>
<td>
The usual <code>+</code> operator is used.
For example
<code>val1 + val2</code>
</td>
</tr>
<tr>
<td>Subtraction</td>
<td>
The usual <code>-</code> operator is used.
For example
<code>val1 - val2</code>
</td>
</tr>
<tr>
<td>Multiplication</td>
<td>
The usual <code>*</code> operator is used.
For example
<code>val1 * val2</code>
</td>
</tr>
<tr>
<td>Division</td>
<td>
The usual <code>/</code> operator is used, or one can use the <code>div</code> operator.
For example
<code>val1 / val2</code>
or
<code>val1 div val2</code>
</td>
</tr>
<tr>
<td>Modulus (or remainder)</td>
<td>
The <code>%</code> operator is used. An alternative is the <code>mod</code>
operator.
For example
<code>5 mod 2</code> gives 1 and is equivalent to <code>5 % 2</code>
</td>
</tr>
<tr>
<td>Side-effect operators</td>
<td>
Some operators exist in side-effect forms.
Their default behavior is to execute the operator and assign the left-hand side with the result.
For instance <code>a += 2</code> is equivalent to <code>a = a + 2</code>
The list of operators is:
<ul>
<li>+=</li>
<li>-=</li>
<li>*=</li>
<li>/=</li>
<li>%=</li>
<li>&amp;=</li>
<li>|=</li>
<li>^=</li>
</ul>
</td>
</tr>
<tr>
<td>Negation</td>
<td>
The unary <code>-</code> operator is used.
For example
<code>-12</code>
</td>
</tr>
</table>
</section>
<section name="Access">
<table>
<tr>
<th width="15%">Operator</th>
<th>Description</th>
</tr>
<tr>
<td>Array access</td>
<td>
Array elements may be accessed using either square brackets or a dotted numeral, e.g.
<code>arr1[0]</code> and <code>arr1.0</code> are equivalent
</td>
</tr>
<tr>
<td>List access</td>
<td>
List elements may be accessed using either square brackets or a dotted numeral, e.g.
<code>list[0]</code> and <code>list.0</code> are equivalent
</td>
</tr>
<tr>
<td>Map access</td>
<td>
Map elements are accessed using square brackets, e.g.
<code>map[0]; map['name']; map[var];</code>
Note that <code>map['7']</code> and <code>map[7]</code> refer to different elements.
Map elements with a numeric key may also be accessed using a dotted numeral, e.g.
<code>map[0]</code> and <code>map.0</code> are equivalent.
<p>Note that <code>map.1</code> and <code>map.01</code> refer to different elements,
while <code>map.1</code> and <code>map[01]</code> are equivalent.</p>
</td>
</tr>
<tr>
<td>JavaBean property access</td>
<td>
Properties of JavaBean objects that define appropriate getter methods can be accessed
using either square brackets or a dotted numeral, e.g.
<code>foo['bar']</code> and <code>foo.bar</code> are equivalent.
The appropriate <code>Foo.getBar()</code> method will be called.
<p>Note that both <code>foo.Bar</code> and <code>foo.bar</code> can be used</p>
</td>
</tr>
<tr>
<td>Indexed JavaBean property access</td>
<td>
Indexed properties of JavaBean objects that define appropriate getter methods can be accessed
using either square brackets or a dotted numeral, e.g.
<code>x.attribute['name']</code> and <code>x.attribute.name</code> are equivalent.
The appropriate <code>Foo.getAttribute(String index)</code> method will be called
</td>
</tr>
<tr>
<td>Public field access</td>
<td>
Public fields of java objects can be accessed using either square brackets or a dotted numeral, e.g.
<code>foo['bar']</code> and <code>foo.bar</code> are equivalent.
</td>
</tr>
<tr>
<td>Duck-typed collection property access</td>
<td>
Properties of Java classes that define public <code>Object get(String name)</code> method can be accessed
using either square brackets or a dotted numeral, e.g.
<code>foo['bar']</code> and <code>foo.bar</code> are equivalent.
The appropriate <code>Foo.get(String index)</code> method will be called with the argument of <code>"bar"</code> String
</td>
</tr>
</table>
</section>
<section name="Conditional">
<table>
<tr>
<th width="15%">Statement</th>
<th>Description</th>
</tr>
<tr>
<td>if</td>
<td>
Classic, if/else statement, e.g.
<code>if ((x * 2) == 5) {
y = 1;
} else {
y = 2;
}</code>
</td>
</tr>
<tr>
<td>for</td>
<td>
<p>Loop through items of an Array, Collection, Map, Iterator or Enumeration, e.g.
<code>for (item : list) {
x = x + item;
}</code>
Where <code>item</code> is a context variable.</p>
<p>The following syntax is also supported:
<code>for (var item : list) {
x = x + item;
}</code>
Where <code>item</code> is a local variable.</p>
<p>Note that <code>item</code> variable is accessible after loop evaluation</p>
<p>The JEXL 1.1 syntax using <code>foreach(item in list)</code> is now <strong>unsupported</strong>.</p>
</td>
</tr>
<tr>
<td>while</td>
<td>
Loop until a condition is satisfied, e.g.
<code>while (x lt 10) {
x = x + 2;
}</code>
</td>
</tr>
<tr>
<td>continue</td>
<td>
Within loops (while/for), allows to skip to the next iteration.
</td>
</tr>
<tr>
<td>break</td>
<td>
Allows to break from a loop (while/for) inconditionally.
</td>
</tr>
</table>
</section>
</body>
</document>