<chapter id="language" xreflabel="The AspectJ Language">
<title>The AspectJ Language</title>
<sect1 id="language-intro">
<title>Introduction</title>
<para>
The previous chapter, <xref linkend="starting" />, was a brief
overview of the AspectJ language. You should read this chapter to
understand AspectJ's syntax and semantics. It covers the same
material as the previous chapter, but more completely and in much
more detail.
</para>
<para>
We will start out by looking at an example aspect that we'll build
out of a pointcut, an introduction, and two pieces of advice. This
example aspect will gives us something concrete to talk about.
</para>
</sect1>
<!-- ============================== -->
<sect1 id="language-anatomy">
<title>The Anatomy of an Aspect</title>
<para>
This lesson explains the parts of AspectJ's aspects. By reading this
lesson you will have an overview of what's in an aspect and you will
be exposed to the new terminology introduced by AspectJ.
</para>
<sect2 id="an-example-aspect" xreflabel="an-example-aspect">
<title>An Example Aspect</title>
<para>
Here's an example of an aspect definition in AspectJ:
</para>
<programlisting><![CDATA[
1 aspect FaultHandler {
2
3 private boolean Server.disabled = false;
4
5 private void reportFault() {
6 System.out.println("Failure! Please fix it.");
7 }
8
9 public static void fixServer(Server s) {
10 s.disabled = false;
11 }
12
13 pointcut services(Server s): target(s) && call(public * *(..));
14
15 before(Server s): services(s) {
16 if (s.disabled) throw new DisabledException();
17 }
18
19 after(Server s) throwing (FaultException e): services(s) {
20 s.disabled = true;
21 reportFault();
22 }
23 }
]]></programlisting>
<para>
The <literal>FaultHandler</literal> consists of one inter-type
field on <literal>Server</literal> (line 03), two methods (lines
05-07 and 09-11), one pointcut definition (line 13), and two pieces
of advice (lines 15-17 and 19-22).
</para>
<para>
This covers the basics of what aspects can contain. In general,
aspects consist of an association of other program entities,
ordinary variables and methods, pointcut definitions, inter-type declarations,
and advice, where advice may be before, after or around advice. The
remainder of this lesson focuses on those crosscut-related
constructs.
</para>
</sect2>
<sect2 id="pointcuts" xreflabel="pointcuts">
<title>Pointcuts</title>
<para>
AspectJ's pointcut definitions give names to pointcuts. Pointcuts
themselves pick out join points, i.e. interesting points in the
execution of a program. These join points can be method or
constructor invocations and executions, the handling of exceptions,
field assignments and accesses, etc. Take, for example, the
pointcut definition in line 13:
</para>
<programlisting><![CDATA[
pointcut services(Server s): target(s) && call(public * *(..))
]]></programlisting>
<para>
This pointcut, named <literal>services</literal>, picks out those
points in the execution of the program when
<literal>Server</literal> objects have their public methods called.
It also allows anyone using the <literal>services</literal>
pointcut to access the <literal>Server</literal> object whose
method is being called.
</para>
<para>
The idea behind this pointcut in the
<literal>FaultHandler</literal> aspect is that
fault-handling-related behavior must be triggered on the calls to
public methods. For example, the server may be unable to proceed
with the request because of some fault. The calls of those methods
are, therefore, interesting events for this aspect, in the sense
that certain fault-related things will happen when these events
occur.
</para>
<para>
Part of the context in which the events occur is exposed by the
formal parameters of the pointcut. In this case, that consists of
objects of type <literal>Server</literal>. That formal parameter
is then being used on the right hand side of the declaration in
order to identify which events the pointcut refers to. In this
case, a pointcut picking out join points where a Server is the
target of some operation (target(s)) is being composed
(<literal><![CDATA[&&]]></literal>, meaning and) with a pointcut
picking out call join points (call(...)). The calls are identified
by signatures that can include wild cards. In this case, there are
wild cards in the return type position (first *), in the name
position (second *) and in the argument list position (..); the
only concrete information is given by the qualifier
<literal>public</literal>.
</para>
<para>
Pointcuts pick out arbitrarily large numbers of join points of a
program. But they pick out only a small number of
<emphasis>kinds</emphasis> of join points. Those kinds of join
points correspond to some of the most important concepts in
Java. Here is an incomplete list: method call, method execution,
exception handling, instantiation, constructor execution, and
field access. Each kind of join point can be picked out by its
own specialized pointcut that you will learn about in other parts
of this guide.
</para>
</sect2>
<!-- ============================== -->
<!-- ============================== -->
<sect2 id="advice" xreflabel="advice">
<title>Advice</title>
<para>
A piece of advice brings together a pointcut and a body of code to
define aspect implementation that runs at join points picked out by
the pointcut. For example, the advice in lines 15-17 specifies that
the following piece of code
</para>
<programlisting><![CDATA[
{
if (s.disabled) throw new DisabledException();
}
]]></programlisting>
<para>
is executed when instances of the <literal>Server</literal> class
have their public methods called, as specified by the pointcut
<literal>services</literal>. More specifically, it runs when those
calls are made, just before the corresponding methods are executed.
</para>
<para>
The advice in lines 19-22 defines another piece of implementation
that is executed on the same pointcut:
</para>
<programlisting><![CDATA[
{
s.disabled = true;
reportFault();
}
]]></programlisting>
<para>
But this second method executes after those operations throw
exception of type <literal>FaultException</literal>.
</para>
<para>
There are two other variations of after advice: upon successful
return and upon return, either successful or with an exception.
There is also a third kind of advice called around. You will see
those in other parts of this guide.
</para>
</sect2>
</sect1>
<!-- ============================== -->
<sect1 id="language-joinPoints">
<title>Join Points and Pointcuts</title>
<para>
Consider the following Java class:
</para>
<programlisting><![CDATA[
class Point {
private int x, y;
Point(int x, int y) { this.x = x; this.y = y; }
void setX(int x) { this.x = x; }
void setY(int y) { this.y = y; }
int getX() { return x; }
int getY() { return y; }
}
]]></programlisting>
<para>
In order to get an intuitive understanding of AspectJ's join points
and pointcuts, let's go back to some of the basic principles of
Java. Consider the following a method declaration in class Point:
</para>
<programlisting><![CDATA[
void setX(int x) { this.x = x; }
]]></programlisting>
<para>
This piece of program says that when method named
<literal>setX</literal> with an <literal>int</literal> argument
called on an object of type <literal>Point</literal>, then the method
body <literal>{ this.x = x; }</literal> is executed. Similarly, the
constructor of the class states that when an object of type
<literal>Point</literal> is instantiated through a constructor with
two <literal>int</literal> arguments, then the constructor body
<literal>{ this.x = x; this.y = y; }</literal> is executed.
</para>
<para>
One pattern that emerges from these descriptions is
<blockquote>
When something happens, then something gets executed.
</blockquote>
In object-oriented programs, there are several kinds of "things that
happen" that are determined by the language. We call these the join
points of Java. Join points consist of things like method calls,
method executions, object instantiations, constructor executions,
field references and handler executions. (See the <xref
linkend="quick" /> for a complete listing.)
</para>
<para>
Pointcuts pick out these join points. For example, the pointcut
</para>
<programlisting><![CDATA[
pointcut setter(): target(Point) &&
(call(void setX(int)) ||
call(void setY(int)));
]]></programlisting>
<para>
picks out each call to <literal>setX(int)</literal> or
<literal>setY(int)</literal> when called on an instance of
<literal>Point</literal>. Here's another example:
</para>
<programlisting><![CDATA[
pointcut ioHandler(): within(MyClass) && handler(IOException);
]]></programlisting>
<para>
This pointcut picks out each the join point when exceptions of type
<literal>IOException</literal> are handled inside the code defined by
class <literal>MyClass</literal>.
</para>
<para>
Pointcut definitions consist of a left-hand side and a right-hand side,
separated by a colon. The left-hand side consists of the pointcut name
and the pointcut parameters (i.e. the data available when the events
happen). The right-hand side consists of the pointcut itself.
</para>
<sect2 id="some-example-pointcuts" xreflabel="some-example-pointcuts">
<title>Some Example Pointcuts</title>
<para>
Here are examples of pointcuts picking out
</para>
<variablelist>
<varlistentry>
<term>when a particular method body executes</term>
<listitem>
<para>
<literal>execution(void Point.setX(int))</literal>
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>when a method is called</term>
<listitem>
<para>
<literal>call(void Point.setX(int))</literal>
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>when an exception handler executes</term>
<listitem>
<para>
<literal>handler(ArrayOutOfBoundsException)</literal>
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
when the object currently executing
(i.e. <literal>this</literal>) is of type
<literal>SomeType</literal>
</term>
<listitem>
<para>
<literal>this(SomeType)</literal>
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
when the target object is of type <literal>SomeType</literal>
</term>
<listitem>
<para>
<literal>target(SomeType)</literal>
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
when the executing code belongs to
class <literal>MyClass</literal>
</term>
<listitem>
<para>
<literal>within(MyClass)</literal>
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>
when the join point is in the control flow of a call to a
<literal>Test</literal>'s no-argument <literal>main</literal>
method
</term>
<listitem>
<para>
<literal>cflow(call(void Test.main()))</literal>
</para>
</listitem>
</varlistentry>
</variablelist>
<para>
Pointcuts compose through the operations <literal>or</literal>
("<literal>||</literal>"), <literal>and</literal>
("<literal><![CDATA[&&]]></literal>") and <literal>not</literal>
("<literal>!</literal>").
</para>
<itemizedlist>
<listitem>
<para>
It is possible to use wildcards. So
<orderedlist>
<listitem>
<para>
<literal>execution(* *(..))</literal>
</para>
</listitem>
<listitem>
<para>
<literal>call(* set(..))</literal>
</para>
</listitem>
</orderedlist>
means (1) the execution of any method regardless of return or
parameter types, and (2) the call to any method named
<literal>set</literal> regardless of return or parameter types
-- in case of overloading there may be more than one such
<literal>set</literal> method; this pointcut picks out calls to
all of them.
</para>
</listitem>
<listitem>
<para>
You can select elements based on types. For example,
<orderedlist>
<listitem>
<para>
<literal>execution(int *())</literal>
</para>
</listitem>
<listitem>
<para>
<literal>call(* setY(long))</literal>
</para>
</listitem>
<listitem>
<para>
<literal>call(* Point.setY(int))</literal>
</para>
</listitem>
<listitem>
<para>
<literal>call(*.new(int, int))</literal>
</para>
</listitem>
</orderedlist>
means (1) the execution of any method with no parameters that
returns an <literal>int</literal>, (2) the call to any
<literal>setY</literal> method that takes a
<literal>long</literal> as an argument, regardless of return
type or declaring type, (3) the call to any of
<literal>Point</literal>'s <literal>setY</literal> methods that
take an <literal>int</literal> as an argument, regardless of
return type, and (4) the call to any classes' constructor, so
long as it takes exactly two <literal>int</literal>s as
arguments.
</para>
</listitem>
<listitem>
<para>
You can compose pointcuts. For example,
<orderedlist>
<listitem>
<para>
<literal>target(Point) <![CDATA[&&]]> call(int *())</literal>
</para>
</listitem>
<listitem>
<para>
<literal>call(* *(..)) <![CDATA[&&]]> (within(Line) || within(Point))</literal>
</para>
</listitem>
<listitem>
<para>
<literal>within(*) <![CDATA[&&]]> execution(*.new(int))</literal>
</para>
</listitem>
<listitem>
<para>
<literal>
!this(Point) <![CDATA[&&]]> call(int *(..))
</literal>
</para>
</listitem>
</orderedlist>
means (1) any call to an <literal>int</literal> method with no
arguments on an instance of <literal>Point</literal>,
regardless of its name, (2) any call to any method where the
call is made from the code in <literal>Point</literal>'s or
<literal>Line</literal>'s type declaration, (3) the execution of
any constructor taking exactly one <literal>int</literal>
argument, regardless of where the call is made from, and
(4) any method call to an <literal>int</literal> method when
the executing object is any type except <literal>Point</literal>.
</para>
</listitem>
<listitem>
<para>
You can select methods and constructors based on their
modifiers and on negations of modifiers. For example, you can
say:
<orderedlist>
<listitem>
<para>
<literal>call(public * *(..))</literal>
</para>
</listitem>
<listitem>
<para>
<literal>execution(!static * *(..))</literal>
</para>
</listitem>
<listitem>
<para>
<literal> execution(public !static * *(..))</literal>
</para>
</listitem>
</orderedlist>
which means (1) any call to a public method, (2) any
execution of a non-static method, and (3) any execution of a
public, non-static method.
</para>
</listitem>
<listitem>
<para>
Pointcuts can also deal with interfaces. For example, given the
interface </para>
<programlisting><![CDATA[
interface MyInterface { ... }
]]></programlisting>
<para>
the pointcut <literal>call(* MyInterface.*(..))</literal> picks
out any call to a method in <literal>MyInterface</literal>'s
signature -- that is, any method defined by
<literal>MyInterface</literal> or inherited by one of its a
supertypes.
</para>
</listitem>
</itemizedlist>
</sect2>
<sect2 id="call-vs-execution" xreflabel="call-vs-execution">
<title>call vs. execution</title>
<para>
When methods and constructors run, there are two interesting times
associated with them. That is when they are called, and when they
actually execute.
</para>
<para>
AspectJ exposes these times as call and execution join points,
respectively, and allows them to be picked out specifically by
<literal>call</literal> and <literal>execution</literal> pointcuts.
</para>
<para>
So what's the difference between these join points? Well, there are a
number of differences:
</para>
<para>
Firstly, the lexical pointcut declarations
<literal>within</literal> and <literal>withincode</literal> match
differently. At a call join point, the enclosing code is that of
the call site. This means that <literal>call(void m())
<![CDATA[&&]]> withincode(void m())</literal> will only capture
directly recursive calls, for example. At an execution join point,
however, the program is already executing the method, so the
enclosing code is the method itself: <literal>execution(void m())
<![CDATA[&&]]> withincode(void m())</literal> is the same as
<literal>execution(void m())</literal>.
</para>
<para>
Secondly, the call join point does not capture super calls to
non-static methods. This is because such super calls are different in
Java, since they don't behave via dynamic dispatch like other calls to
non-static methods.
</para>
<para>
The rule of thumb is that if you want to pick a join point that
runs when an actual piece of code runs (as is often the case for
tracing), use <literal>execution</literal>, but if you want to pick
one that runs when a particular <emphasis>signature</emphasis> is
called (as is often the case for production aspects), use
<literal>call</literal>.
</para>
</sect2>
<!-- ============================== -->
<sect2 id="pointcut-composition" xreflabel="pointcut-composition">
<title>Pointcut composition</title>
<para>
Pointcuts are put together with the operators and (spelled
<literal>&&</literal>), or (spelled <literal>||</literal>),
and not (spelled <literal>!</literal>). This allows the creation
of very powerful pointcuts from the simple building blocks of
primitive pointcuts. This composition can be somewhat confusing
when used with primitive pointcuts like <literal>cflow</literal>
and <literal>cflowbelow</literal>. Here's an example:
</para>
<para>
<literal>cflow(<replaceable>P</replaceable>)</literal> picks out
each join point in the control flow of the join points picked out
by <replaceable>P</replaceable>. So, pictorially:
</para>
<programlisting>
P ---------------------
\
\ cflow of P
\
</programlisting>
<para>
What does <literal>cflow(<replaceable>P</replaceable>) &&
cflow(<replaceable>Q</replaceable>)</literal> pick out? Well, it
picks out each join point that is in both the control flow of
<replaceable>P</replaceable> and in the control flow of
<replaceable>Q</replaceable>. So...
</para>
<programlisting>
P ---------------------
\
\ cflow of P
\
\
\
Q -------------\-------
\ \
\ cflow of Q \ cflow(P) && cflow(Q)
\ \
</programlisting>
<para>
Note that <replaceable>P</replaceable> and
<replaceable>Q</replaceable> might not have any join points in
common... but their control flows might have join points in common.
</para>
<para>
But what does <literal>cflow(<replaceable>P</replaceable>
&& <replaceable>Q</replaceable>)</literal> mean? Well, it
means the control flow of those join points that are both picked
out by <replaceable>P</replaceable> and picked out by
<replaceable>Q</replaceable>.
</para>
<programlisting>
P && Q -------------------
\
\ cflow of (P && Q)
\
</programlisting>
<para>
and if there are <emphasis>no</emphasis> join points that are both
picked by <replaceable>P</replaceable> and picked out by
<replaceable>Q</replaceable>, then there's no chance that there are
any join points in the control flow of
<literal>(<replaceable>P</replaceable> &&
<replaceable>Q</replaceable>)</literal>.
</para>
<para>
Here's some code that expresses this.
</para>
<programlisting><![CDATA[
public class Test {
public static void main(String[] args) {
foo();
}
static void foo() {
goo();
}
static void goo() {
System.out.println("hi");
}
}
aspect A {
pointcut fooPC(): execution(void Test.foo());
pointcut gooPC(): execution(void Test.goo());
pointcut printPC(): call(void java.io.PrintStream.println(String));
before(): cflow(fooPC()) && cflow(gooPC()) && printPC() && !within(A) {
System.out.println("should occur");
}
before(): cflow(fooPC() && gooPC()) && printPC() && !within(A) {
System.out.println("should not occur");
}
}
]]></programlisting>
<para>
The <literal>!within(<replaceable>A</replaceable>)</literal>
pointcut above is required to avoid the <literal>printPC</literal>
pointcut applying to the <literal>System.out.println</literal>
call in the advice body. If this was not present a recursive call
would result as the pointcut would apply to its own advice.
(See <xref linkend="pitfalls-infiniteLoops"/> for more details.)
</para>
</sect2>
<!-- ============================== -->
<sect2 id="pointcut-parameters" xreflabel="pointcut-parameters">
<title>Pointcut Parameters</title>
<para>
Consider again the first pointcut definition in this chapter:
</para>
<programlisting><![CDATA[
pointcut setter(): target(Point) &&
(call(void setX(int)) ||
call(void setY(int)));
]]></programlisting>
<para>
As we've seen, this pointcut picks out each call to
<literal>setX(int)</literal> or <literal>setY(int)</literal>
methods where the target is an instance of
<literal>Point</literal>. The pointcut is given the name
<literal>setters</literal> and no parameters on the left-hand
side. An empty parameter list means that none of the context from
the join points is published from this pointcut. But consider
another version of version of this pointcut definition:
</para>
<programlisting><![CDATA[
pointcut setter(Point p): target(p) &&
(call(void setX(int)) ||
call(void setY(int)));
]]></programlisting>
<para>
This version picks out exactly the same join points. But in this
version, the pointcut has one parameter of type
<literal>Point</literal>. This means that any advice that uses this
pointcut has access to a <literal>Point</literal> from each join
point picked out by the pointcut. Inside the pointcut definition
this <literal>Point</literal> is named <literal>p</literal> is
available, and according to the right-hand side of the definition,
that <literal>Point p</literal> comes from the
<literal>target</literal> of each matched join point.
</para>
<para>
Here's another example that illustrates the flexible mechanism for
defining pointcut parameters:
</para>
<programlisting><![CDATA[
pointcut testEquality(Point p): target(Point) &&
args(p) &&
call(boolean equals(Object));
]]></programlisting>
<para>
This pointcut also has a parameter of type
<literal>Point</literal>. Similar to the
<literal>setters</literal> pointcut, this means that anyone using
this pointcut has access to a <literal>Point</literal> from each
join point. But in this case, looking at the right-hand side we
find that the object named in the parameters is not the target
<literal>Point</literal> object that receives the call; it's the
argument (also of type <literal>Point</literal>) passed to the
<literal>equals</literal> method when some other
<literal>Point</literal> is the target. If we wanted access to both
<literal>Point</literal>s, then the pointcut definition that would
expose target <literal>Point p1</literal> and argument
<literal>Point p2</literal> would be
</para>
<programlisting><![CDATA[
pointcut testEquality(Point p1, Point p2): target(p1) &&
args(p2) &&
call(boolean equals(Object));
]]></programlisting>
<para>
Let's look at another variation of the <literal>setters</literal> pointcut:
</para>
<programlisting><![CDATA[
pointcut setter(Point p, int newval): target(p) &&
args(newval) &&
(call(void setX(int)) ||
call(void setY(int)));
]]></programlisting>
<para>
In this case, a <literal>Point</literal> object and an
<literal>int</literal> value are exposed by the named
pointcut. Looking at the the right-hand side of the definition, we
find that the <literal>Point</literal> object is the target object,
and the <literal>int</literal> value is the called method's
argument.
</para>
<para>
The use of pointcut parameters is relatively flexible. The most
important rule is that all the pointcut parameters must be bound at
every join point picked out by the pointcut. So, for example, the
following pointcut definition will result in a compilation error:
<programlisting><![CDATA[
pointcut badPointcut(Point p1, Point p2):
(target(p1) && call(void setX(int))) ||
(target(p2) && call(void setY(int)));
]]></programlisting>
because <literal>p1</literal> is only bound when calling
<literal>setX</literal>, and <literal>p2</literal> is only bound
when calling <literal>setY</literal>, but the pointcut picks out
all of these join points and tries to bind both
<literal>p1</literal> and <literal>p2</literal>.
</para>
</sect2>
<!-- ============================== -->
<sect2 id="example" xreflabel="example">
<title>Example: <literal>HandleLiveness</literal></title>
<para>
The example below consists of two object classes (plus an exception
class) and one aspect. Handle objects delegate their public,
non-static operations to their <literal>Partner</literal>
objects. The aspect <literal>HandleLiveness</literal> ensures that,
before the delegations, the partner exists and is alive, or else it
throws an exception.
</para>
<programlisting><![CDATA[
class Handle {
Partner partner = new Partner();
public void foo() { partner.foo(); }
public void bar(int x) { partner.bar(x); }
public static void main(String[] args) {
Handle h1 = new Handle();
h1.foo();
h1.bar(2);
}
}
class Partner {
boolean isAlive() { return true; }
void foo() { System.out.println("foo"); }
void bar(int x) { System.out.println("bar " + x); }
}
aspect HandleLiveness {
before(Handle handle): target(handle) && call(public * *(..)) {
if ( handle.partner == null || !handle.partner.isAlive() ) {
throw new DeadPartnerException();
}
}
}
class DeadPartnerException extends RuntimeException {}
]]></programlisting>
</sect2>
<sect2 id="pointcut-best-practice" xreflabel="pointcut-best-practice">
<title>Writing good pointcuts</title>
<para>
During compilation, AspectJ processes pointcuts in order to try and optimize matching performance. Examining code and determining
if each join point matches (statically or dynamically) a given pointcut is a costly process.
(A dynamic match means the match cannot be fully determined from static analysis and a test will be placed in the code
to determine if there is an actual match when the code is running).
On first encountering a pointcut declaration, AspectJ will rewrite it into an optimal form for the matching process.
What does this mean? Basically pointcuts are rewritten in DNF (Disjunctive Normal Form) and the components of the pointcut
are sorted such that those components that are cheaper to evaluate are checked first. This means users do not have to worry
about understanding the performance of various pointcut designators and may supply them in any order in their
pointcut declarations.
</para>
<para>
However, AspectJ can only work with what it is told, and for optimal performance of matching the user should think
about what they are trying to achieve and narrow the search space for matches as much as they can in the definition.
Basically there are three kinds of pointcut designator: kinded, scoping and context:
</para>
<itemizedlist>
<listitem>
Kinded designators are those which select a particular kind of join point. For example: execution, get, set, call, handler
</listitem>
<listitem>
Scoping designators are those which select a group of join points of interest (of probably many kinds). For example: within, withincode
</listitem>
<listitem>
Contextual designators are those that match (and optionally bind) based on context. For example: this, target, @annotation
</listitem>
</itemizedlist>
<para>
A well written pointcut should
try and include at least the first two types (kinded and scoping), whilst the contextual designators may be included if wishing to
match based on join point context, or bind that context for use in the advice. Supplying either just a kinded designator or
just a contextual designator will work but could affect weaving performance (time and memory used)
due to all the extra processing and analysis.
Scoping designators are very fast to match, they can very quickly dismiss groups of join points that should not be further
processed - that is why a good pointcut should always include one if possible.
</para>
</sect2>
</sect1>
<!-- ============================== -->
<sect1 id="language-advice">
<title>Advice</title>
<para>
Advice defines pieces of aspect implementation that execute at
well-defined points in the execution of the program. Those points can
be given either by named pointcuts (like the ones you've seen above)
or by anonymous pointcuts. Here is an example of an advice on a named
pointcut:
</para>
<programlisting><![CDATA[
pointcut setter(Point p1, int newval): target(p1) && args(newval)
(call(void setX(int) ||
call(void setY(int)));
before(Point p1, int newval): setter(p1, newval) {
System.out.println("About to set something in " + p1 +
" to the new value " + newval);
}
]]></programlisting>
<para>
And here is exactly the same example, but using an anonymous
pointcut:
</para>
<programlisting><![CDATA[
before(Point p1, int newval): target(p1) && args(newval)
(call(void setX(int)) ||
call(void setY(int))) {
System.out.println("About to set something in " + p1 +
" to the new value " + newval);
}
]]></programlisting>
<para>
Here are examples of the different advice:
</para>
<para>
This before advice runs just before the join points picked out by the
(anonymous) pointcut:
</para>
<programlisting><![CDATA[
before(Point p, int x): target(p) && args(x) && call(void setX(int)) {
if (!p.assertX(x)) return;
}
]]></programlisting>
<para>
This after advice runs just after each join point picked out by the
(anonymous) pointcut, regardless of whether it returns normally or throws
an exception:
</para>
<programlisting><![CDATA[
after(Point p, int x): target(p) && args(x) && call(void setX(int)) {
if (!p.assertX(x)) throw new PostConditionViolation();
}
]]></programlisting>
<para>
This after returning advice runs just after each join point picked
out by the (anonymous) pointcut, but only if it returns normally.
The return value can be accessed, and is named <literal>x</literal>
here. After the advice runs, the return value is returned:
</para>
<programlisting><![CDATA[
after(Point p) returning(int x): target(p) && call(int getX()) {
System.out.println("Returning int value " + x + " for p = " + p);
}
]]></programlisting>
<para>
This after throwing advice runs just after each join point picked out by
the (anonymous) pointcut, but only when it throws an exception of type
<literal>Exception</literal>. Here the exception value can be accessed
with the name <literal>e</literal>. The advice re-raises the exception
after it's done:
</para>
<programlisting><![CDATA[
after() throwing(Exception e): target(Point) && call(void setX(int)) {
System.out.println(e);
}
]]></programlisting>
<para>
This around advice traps the execution of the join point; it runs
<emphasis>instead</emphasis> of the join point. The original action
associated with the join point can be invoked through the special
<literal>proceed</literal> call:
</para>
<programlisting><![CDATA[
void around(Point p, int x): target(p)
&& args(x)
&& call(void setX(int)) {
if (p.assertX(x)) proceed(p, x);
p.releaseResources();
}
]]></programlisting>
</sect1>
<!-- ============================== -->
<sect1 id="language-interType">
<title>Inter-type declarations</title>
<para>
Aspects can declare members (fields, methods, and constructors) that
are owned by other types. These are called inter-type members.
Aspects can also declare that other types implement new interfaces or
extend a new class. Here are examples of some such inter-type
declarations:
</para>
<para>
This declares that each <literal>Server</literal> has a
<literal>boolean</literal> field named <literal>disabled</literal>,
initialized to <literal>false</literal>:
<programlisting><![CDATA[
private boolean Server.disabled = false;
]]></programlisting>
It is declared <literal>private</literal>, which means that it is
private <emphasis>to the aspect</emphasis>: only code in the aspect
can see the field. And even if <literal>Server</literal> has
another private field named <literal>disabled</literal> (declared in
<literal>Server</literal> or in another aspect) there won't be a name
collision, since no reference to <literal>disabled</literal> will be
ambiguous.
</para>
<para>
This declares that each <literal>Point</literal> has an
<literal>int</literal> method named <literal>getX</literal> with no
arguments that returns whatever <literal>this.x</literal> is:
<programlisting><![CDATA[
public int Point.getX() { return this.x; }
]]></programlisting>
Inside the body, <literal>this</literal> is the
<literal>Point</literal> object currently executing. Because the
method is publically declared any code can call it, but if there is
some other <literal>Point.getX()</literal> declared there will be a
compile-time conflict.
</para>
<para>
This publically declares a two-argument constructor for
<literal>Point</literal>:
<programlisting><![CDATA[
public Point.new(int x, int y) { this.x = x; this.y = y; }
]]></programlisting>
</para>
<para>
This publicly declares that each <literal>Point</literal> has an
<literal>int</literal> field named <literal>x</literal>, initialized
to zero:
<programlisting><![CDATA[
public int Point.x = 0;
]]></programlisting>
Because this is publically declared, it is an error if
<literal>Point</literal> already has a field named
<literal>x</literal> (defined by <literal>Point</literal> or by
another aspect).
</para>
<para>
This declares that the <literal>Point</literal> class implements the
<literal>Comparable</literal> interface:
<programlisting><![CDATA[
declare parents: Point implements Comparable;
]]></programlisting>
Of course, this will be an error unless <literal>Point</literal>
defines the methods required by <literal>Comparable</literal>.
</para>
<para>
This declares that the <literal>Point</literal> class extends the
<literal>GeometricObject</literal> class.
<programlisting><![CDATA[
declare parents: Point extends GeometricObject;
]]></programlisting>
</para>
<para>
An aspect can have several inter-type declarations. For example, the
following declarations
<programlisting><![CDATA[
public String Point.name;
public void Point.setName(String name) { this.name = name; }
]]></programlisting>
publicly declare that Point has both a String field
<literal>name</literal> and a <literal>void</literal> method
<literal>setName(String)</literal> (which refers to the
<literal>name</literal> field declared by the aspect).
</para>
<para>
An inter-type member can only have one target type, but often you may
wish to declare the same member on more than one type. This can be
done by using an inter-type member in combination with a private
interface:
<programlisting><![CDATA[
aspect A {
private interface HasName {}
declare parents: (Point || Line || Square) implements HasName;
private String HasName.name;
public String HasName.getName() { return name; }
}
]]></programlisting>
This declares a marker interface <literal>HasName</literal>, and also declares that any
type that is either <literal>Point</literal>,
<literal>Line</literal>, or <literal>Square</literal> implements that
interface. It also privately declares that all <literal>HasName</literal>
object have a <literal>String</literal> field called
<literal>name</literal>, and publically declares that all
<literal>HasName</literal> objects have a <literal>String</literal>
method <literal>getName()</literal> (which refers to the privately
declared <literal>name</literal> field).
</para>
<para>
As you can see from the above example, an aspect can declare that
interfaces have fields and methods, even non-constant fields and
methods with bodies.
</para>
<!-- ============================== -->
<sect2 id="inter-type-scope" xreflabel="inter-type-scope">
<title>Inter-type Scope</title>
<para>
AspectJ allows private and package-protected (default) inter-type declarations in
addition to public inter-type declarations. Private means private in
relation to the aspect, not necessarily the target type. So, if an
aspect makes a private inter-type declaration of a field
<programlisting><![CDATA[
private int Foo.x;
]]></programlisting>
Then code in the aspect can refer to <literal>Foo</literal>'s
<literal>x</literal> field, but nobody else can. Similarly, if an
aspect makes a package-protected introduction,
</para>
<programlisting><![CDATA[
int Foo.x;
]]></programlisting>
<para>
then everything in the aspect's package (which may or may not be
<literal>Foo</literal>'s package) can access <literal>x</literal>.
</para>
</sect2>
<!-- ============================== -->
<sect2 id="example-pointassertions" xreflabel="example-pointassertions">
<title>Example: <literal>PointAssertions</literal></title>
<para>
The example below consists of one class and one aspect. The aspect
privately declares the assertion methods of
<literal>Point</literal>, <literal>assertX</literal> and
<literal>assertY</literal>. It also guards calls to
<literal>setX</literal> and <literal>setY</literal> with calls to
these assertion methods. The assertion methods are declared
privately because other parts of the program (including the code in
<literal>Point</literal>) have no business accessing the assert
methods. Only the code inside of the aspect can call those
methods.
</para>
<programlisting><![CDATA[
class Point {
int x, y;
public void setX(int x) { this.x = x; }
public void setY(int y) { this.y = y; }
public static void main(String[] args) {
Point p = new Point();
p.setX(3); p.setY(333);
}
}
aspect PointAssertions {
private boolean Point.assertX(int x) {
return (x <= 100 && x >= 0);
}
private boolean Point.assertY(int y) {
return (y <= 100 && y >= 0);
}
before(Point p, int x): target(p) && args(x) && call(void setX(int)) {
if (!p.assertX(x)) {
System.out.println("Illegal value for x"); return;
}
}
before(Point p, int y): target(p) && args(y) && call(void setY(int)) {
if (!p.assertY(y)) {
System.out.println("Illegal value for y"); return;
}
}
}
]]></programlisting>
</sect2>
</sect1>
<!-- ================================================== -->
<sect1 id="language-thisJoinPoint">
<title>thisJoinPoint</title>
<para>
AspectJ provides a special reference variable,
<literal>thisJoinPoint</literal>, that contains reflective
information about the current join point for the advice to use. The
<literal>thisJoinPoint</literal> variable can only be used in the
context of advice, just like <literal>this</literal> can only be used
in the context of non-static methods and variable initializers. In
advice, <literal>thisJoinPoint</literal> is an object of type <ulink
url="../api/org/aspectj/lang/JoinPoint.html"><literal>org.aspectj.lang.JoinPoint</literal></ulink>.
</para>
<para>
One way to use it is simply to print it out. Like all Java objects,
<literal>thisJoinPoint</literal> has a <literal>toString()</literal>
method that makes quick-and-dirty tracing easy:
</para>
<programlisting><![CDATA[
aspect TraceNonStaticMethods {
before(Point p): target(p) && call(* *(..)) {
System.out.println("Entering " + thisJoinPoint + " in " + p);
}
}
]]></programlisting>
<para>
The type of <literal>thisJoinPoint</literal> includes a rich
reflective class hierarchy of signatures, and can be used to access
both static and dynamic information about join points such as the
arguments of the join point:
<programlisting><![CDATA[
thisJoinPoint.getArgs()
]]></programlisting>
In addition, it holds an object consisting of all the static
information about the join point such as corresponding line number
and static signature:
<programlisting><![CDATA[
thisJoinPoint.getStaticPart()
]]></programlisting>
If you only need the static information about the join point, you may
access the static part of the join point directly with the special
variable <literal>thisJoinPointStaticPart</literal>. Using
<literal>thisJoinPointStaticPart</literal> will avoid the run-time
creation of the join point object that may be necessary when using
<literal>thisJoinPoint</literal> directly.
</para>
<para>It is always the case that
</para>
<programlisting><![CDATA[
thisJoinPointStaticPart == thisJoinPoint.getStaticPart()
thisJoinPoint.getKind() == thisJoinPointStaticPart.getKind()
thisJoinPoint.getSignature() == thisJoinPointStaticPart.getSignature()
thisJoinPoint.getSourceLocation() == thisJoinPointStaticPart.getSourceLocation()
]]></programlisting>
<para>
One more reflective variable is available:
<literal>thisEnclosingJoinPointStaticPart</literal>. This, like
<literal>thisJoinPointStaticPart</literal>, only holds the static
part of a join point, but it is not the current but the enclosing
join point. So, for example, it is possible to print out the calling
source location (if available) with
</para>
<programlisting><![CDATA[
before() : execution (* *(..)) {
System.err.println(thisEnclosingJoinPointStaticPart.getSourceLocation())
}
]]></programlisting>
</sect1>
</chapter>