Class pointcut that intercepts a method call with field access - aop

I am writing some aspects to make my gradle pugin development a little cleaner. In gradle, there is an interface, like this
interface Plugin {
def apply(Project project);
}
Applied to a plugin
class MyPlugin implements Plugin<Project> {
def apply(Project project) {
do stuff
}
}
Now, i want to be able to annotate this class like this
#OnlyAllowedOnRoot
class MyPlugin implements Plugin<Project> {
def apply(Project project) {
do stuff
}
}
and have a pointcut that fires when the 'apply' method is fired, and pick up the parameter. because the logic for the pointcut would be
if (project.rootProject.name != project.name) {
throw new GradleScriptExeption("This plugin can only be applied to root")
}
how would I do this? This example is the foundation for about a dozen other pointcuts i would want to write, but i really don't know where to start. I know i can annotate the apply method directly, but im worried about readability, which is mostly the reason why im doing this to begin with. I can if i have to, but id rather not. And because of the lifecycle of Gradle, it must be checked when the 'apply' method is called, it can't be checked at the instantiation.

Here is a complete example, but in Java, not in Groovy. It should not make any difference if you use Groovy, though. BTW, I did not want to add the Gradle API as a dependency to my project, so I just replicated the relevant parts of its API with the right package names and signatures. Not being a Gradle user myself, I implemented the root project property hierarchically as a direct parent, not as an absolute root. If Gradle does it differently, just adjust the condition in the aspect throwing the exception back to your own sample code.
Gradle API:
package org.gradle.api;
public class Project {
private String name;
private Project rootProject;
public Project(String name, Project rootProject) {
this.name = name;
this.rootProject = rootProject;
}
public String getName() {
return name;
}
public Project getRootProject() {
return rootProject;
}
#Override
public String toString() {
return "Project(name = " + name + ", rootProject = " + rootProject + ")";
}
}
package org.gradle.api;
public interface Plugin<T> {
void apply(T target);
}
package org.gradle.api;
public class GradleScriptExeption extends RuntimeException {
private static final long serialVersionUID = 1L;
public GradleScriptExeption(String message, Throwable cause) {
super(message, cause);
}
}
Marker annotation + plugins:
package de.scrum_master.app;
import org.gradle.api.Plugin;
import org.gradle.api.Project;
class NormalPlugin implements Plugin<Project> {
public void apply(Project project) {}
}
package de.scrum_master.app;
import static java.lang.annotation.RetentionPolicy.RUNTIME;
import java.lang.annotation.Retention;
#Retention(RUNTIME)
public #interface OnlyAllowedOnRoot {}
package de.scrum_master.app;
import org.gradle.api.Plugin;
import org.gradle.api.Project;
#OnlyAllowedOnRoot
class RootPlugin implements Plugin<Project> {
public void apply(Project project) {}
}
Driver application:
package de.scrum_master.app;
import org.gradle.api.Project;
public class Application {
public static void main(String[] args) {
Project rootProject = new Project("root", null);
Project childProject = new Project("child", rootProject);
Project grandChildProject = new Project("grandchild", childProject);
NormalPlugin normalPlugin = new NormalPlugin();
normalPlugin.apply(rootProject);
normalPlugin.apply(childProject);
normalPlugin.apply(grandChildProject);
RootPlugin rootPlugin = new RootPlugin();
rootPlugin.apply(rootProject);
rootPlugin.apply(childProject);
rootPlugin.apply(grandChildProject);
}
}
Aspect:
package de.scrum_master.aspect;
import org.aspectj.lang.JoinPoint;
import org.aspectj.lang.annotation.Aspect;
import org.aspectj.lang.annotation.Before;
import org.aspectj.lang.annotation.Pointcut;
import org.gradle.api.GradleScriptExeption;
import org.gradle.api.Plugin;
import org.gradle.api.Project;
#Aspect
public class GradlePluginAspect {
#Pointcut("execution(void apply(*)) && target(plugin) && args(project)")
private static void pluginExecution(Plugin plugin, Project project) {}
#Before("pluginExecution(plugin, project) && #target(de.scrum_master.app.OnlyAllowedOnRoot)")
public void illegalRootPlugin(JoinPoint thisJoinPoint, Plugin plugin, Project project) {
if (project.getRootProject() != null)
throw new GradleScriptExeption("Cannot apply " + plugin.getClass().getSimpleName() + " to non-root project " + project, null);
}
#Before("pluginExecution(plugin, project)")
public void logPluginApply(JoinPoint thisJoinPoint, Plugin plugin, Project project) {
System.out.println("Applying " + plugin.getClass().getSimpleName() + " to " + project);
}
}
Console log:
Applying NormalPlugin to Project(name = root, rootProject = null)
Applying NormalPlugin to Project(name = child, rootProject = Project(name = root, rootProject = null))
Applying NormalPlugin to Project(name = grandchild, rootProject = Project(name = child, rootProject = Project(name = root, rootProject = null)))
Applying RootPlugin to Project(name = root, rootProject = null)
Exception in thread "main" org.gradle.api.GradleScriptExeption: Cannot apply RootPlugin to non-root project Project(name = child, rootProject = Project(name = root, rootProject = null))
at de.scrum_master.aspect.GradlePluginAspect.illegalRootPlugin(GradlePluginAspect.aj:19)
at de.scrum_master.app.RootPlugin.apply(RootPlugin.java:8)
at de.scrum_master.app.Application.main(Application.java:18)

Related

Spring AOP - passing arguments between annotated methods

i've written a utility to monitor individual business transactions. For example, Alice calls a method which calls more methods and i want info on just Alice's call, separate from Bob's call to the same method.
Right now the entry point creates a Transaction object and it's passed as an argument to each method:
class Example {
public Item getOrderEntryPoint(int orderId) {
Transaction transaction = transactionManager.create();
transaction.trace("getOrderEntryPoint");
Order order = getOrder(orderId, transaction);
transaction.stop();
logger.info(transaction);
return item;
}
private Order getOrder(int orderId, Transaction t) {
t.trace("getOrder");
Order order = getItems(itemId, t);
t.addStat("number of items", order.getItems().size());
for (Item item : order.getItems()) {
SpecialOffer offer = getSpecialOffer(item, t);
if (null != offer) {
t.incrementStat("offers", 1);
}
}
t.stop();
return order;
}
private SpecialOffer getSpecialOffer(Item item, Transaction t) {
t.trace("getSpecialOffer(" + item.id + ")", TraceCategory.Database);
return offerRepository.getByItem(item);
t.stop();
}
}
This will print to the log something like:
Transaction started by Alice at 10:42
Statistics:
number of items : 3
offers : 1
Category Timings (longest first):
DB : 2s 903ms
code : 187ms
Timings (longest first):
getSpecialOffer(1013) : 626ms
getItems : 594ms
Trace:
getOrderEntryPoint (7ms)
getOrder (594ms)
getSpecialOffer(911) (90ms)
getSpecialOffer(1013) (626ms)
getSpecialOffer(2942) (113ms)
It works great but passing the transaction object around is ugly. Someone suggested AOP but i don't see how to pass the transaction created in the first method to all the other methods.
The Transaction object is pretty simple:
public class Transaction {
private String uuid = UUID.createRandom();
private List<TraceEvent> events = new ArrayList<>();
private Map<String,Int> stats = new HashMap<>();
}
class TraceEvent {
private String name;
private long durationInMs;
}
The app that uses it is a Web app, and this multi-threaded, but the individual transactions are on a single thread - no multi-threading, async code, competition for resources, etc.
My attempt at an annotation:
#Around("execution(* *(..)) && #annotation(Trace)")
public Object around(ProceedingJoinPoint point) {
String methodName = MethodSignature.class.cast(point.getSignature()).getMethod().getName();
//--- Where do i get this call's instance of TRANSACTION from?
if (null == transaction) {
transaction = TransactionManager.createTransaction();
}
transaction.trace(methodName);
Object result = point.proceed();
transaction.stop();
return result;
Introduction
Unfortunately, your pseudo code does not compile. It contains several syntactical and logical errors. Furthermore, some helper classes are missing. If I did not have spare time today and was looking for a puzzle to solve, I would not have bothered making my own MCVE out of it, because that would actually have been your job. Please do read the MCVE article and learn to create one next time, otherwise you will not get a lot of qualified help here. This was your free shot because you are new on SO.
Original situation: passing through transaction objects in method calls
Application helper classes:
package de.scrum_master.app;
public class Item {
private int id;
public Item(int id) {
this.id = id;
}
public int getId() {
return id;
}
#Override
public String toString() {
return "Item[id=" + id + "]";
}
}
package de.scrum_master.app;
public class SpecialOffer {}
package de.scrum_master.app;
public class OfferRepository {
public SpecialOffer getByItem(Item item) {
if (item.getId() < 30)
return new SpecialOffer();
return null;
}
}
package de.scrum_master.app;
import java.util.ArrayList;
import java.util.List;
public class Order {
private int id;
public Order(int id) {
this.id = id;
}
public List<Item> getItems() {
List<Item> items = new ArrayList<>();
int offset = id == 12345 ? 0 : 1;
items.add(new Item(11 + offset, this));
items.add(new Item(22 + offset, this));
items.add(new Item(33 + offset, this));
return items;
}
}
Trace classes:
package de.scrum_master.trace;
public enum TraceCategory {
Code, Database
}
package de.scrum_master.trace;
class TraceEvent {
private String name;
private TraceCategory category;
private long durationInMs;
private boolean finished = false;
public TraceEvent(String name, TraceCategory category, long startTime) {
this.name = name;
this.category = category;
this.durationInMs = startTime;
}
public long getDurationInMs() {
return durationInMs;
}
public void setDurationInMs(long durationInMs) {
this.durationInMs = durationInMs;
}
public boolean isFinished() {
return finished;
}
public void setFinished(boolean finished) {
this.finished = finished;
}
#Override
public String toString() {
return "TraceEvent[name=" + name + ", category=" + category +
", durationInMs=" + durationInMs + ", finished=" + finished + "]";
}
}
Transaction classes:
Here I tried to mimic your own Transaction class with as few as possible changes, but there was a lot I had to add and modify in order to emulate a simplified version of your trace output. This is not thread-safe and the way I am locating the last unfinished TraceEvent is not nice and only works cleanly if there are not exceptions. But you get the idea, I hope. The point is to just make it basically work and subsequently get log output similar to your example. If this was originally my code, I would have solved it differently.
package de.scrum_master.trace;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import java.util.Map.Entry;
import java.util.UUID;
public class Transaction {
private String uuid = UUID.randomUUID().toString();
private List<TraceEvent> events = new ArrayList<>();
private Map<String, Integer> stats = new HashMap<>();
public void trace(String message) {
trace(message, TraceCategory.Code);
}
public void trace(String message, TraceCategory category) {
events.add(new TraceEvent(message, category, System.currentTimeMillis()));
}
public void stop() {
TraceEvent event = getLastUnfinishedEvent();
event.setDurationInMs(System.currentTimeMillis() - event.getDurationInMs());
event.setFinished(true);
}
private TraceEvent getLastUnfinishedEvent() {
return events
.stream()
.filter(event -> !event.isFinished())
.reduce((first, second) -> second)
.orElse(null);
}
public void addStat(String text, int size) {
stats.put(text, size);
}
public void incrementStat(String text, int increment) {
Integer currentCount = stats.get(text);
if (currentCount == null)
currentCount = 0;
stats.put(text, currentCount + increment);
}
#Override
public String toString() {
return "Transaction {" +
toStringUUID() +
toStringStats() +
toStringEvents() +
"\n}\n";
}
private String toStringUUID() {
return "\n uuid = " + uuid;
}
private String toStringStats() {
String result = "\n stats = {";
for (Entry<String, Integer> statEntry : stats.entrySet())
result += "\n " + statEntry;
return result + "\n }";
}
private String toStringEvents() {
String result = "\n events = {";
for (TraceEvent event : events)
result += "\n " + event;
return result + "\n }";
}
}
package de.scrum_master.trace;
public class TransactionManager {
public Transaction create() {
return new Transaction();
}
}
Example driver application:
package de.scrum_master.app;
import de.scrum_master.trace.TraceCategory;
import de.scrum_master.trace.Transaction;
import de.scrum_master.trace.TransactionManager;
public class Example {
private TransactionManager transactionManager = new TransactionManager();
private OfferRepository offerRepository = new OfferRepository();
public Order getOrderEntryPoint(int orderId) {
Transaction transaction = transactionManager.create();
transaction.trace("getOrderEntryPoint");
sleep(100);
Order order = getOrder(orderId, transaction);
transaction.stop();
System.out.println(transaction);
return order;
}
private Order getOrder(int orderId, Transaction t) {
t.trace("getOrder");
sleep(200);
Order order = new Order(orderId);
t.addStat("number of items", order.getItems().size());
for (Item item : order.getItems()) {
SpecialOffer offer = getSpecialOffer(item, t);
if (null != offer)
t.incrementStat("special offers", 1);
}
t.stop();
return order;
}
private SpecialOffer getSpecialOffer(Item item, Transaction t) {
t.trace("getSpecialOffer(" + item.getId() + ")", TraceCategory.Database);
sleep(50);
SpecialOffer specialOffer = offerRepository.getByItem(item);
t.stop();
return specialOffer;
}
private void sleep(long millis) {
try {
Thread.sleep(millis);
} catch (InterruptedException e) {
e.printStackTrace();
}
}
public static void main(String[] args) {
new Example().getOrderEntryPoint(12345);
new Example().getOrderEntryPoint(23456);
}
}
If you run this code, the output is as follows:
Transaction {
uuid = 62ec9739-bd32-4a56-b6b3-a8a13624961a
stats = {
special offers=2
number of items=3
}
events = {
TraceEvent[name=getOrderEntryPoint, category=Code, durationInMs=561, finished=true]
TraceEvent[name=getOrder, category=Code, durationInMs=451, finished=true]
TraceEvent[name=getSpecialOffer(11), category=Database, durationInMs=117, finished=true]
TraceEvent[name=getSpecialOffer(22), category=Database, durationInMs=69, finished=true]
TraceEvent[name=getSpecialOffer(33), category=Database, durationInMs=63, finished=true]
}
}
Transaction {
uuid = a420cd70-96e5-44c4-a0a4-87e421d05e87
stats = {
special offers=2
number of items=3
}
events = {
TraceEvent[name=getOrderEntryPoint, category=Code, durationInMs=469, finished=true]
TraceEvent[name=getOrder, category=Code, durationInMs=369, finished=true]
TraceEvent[name=getSpecialOffer(12), category=Database, durationInMs=53, finished=true]
TraceEvent[name=getSpecialOffer(23), category=Database, durationInMs=63, finished=true]
TraceEvent[name=getSpecialOffer(34), category=Database, durationInMs=53, finished=true]
}
}
AOP refactoring
Preface
Please note that I am using AspectJ here because two things about your code would never work with Spring AOP because it works with a delegation pattern based on dynamic proxies:
self-invocation (internally calling a method of the same class or super-class)
intercepting private methods
Because of these Spring AOP limitations I advise you to either refactor your code so as to avoid the two issues above or to configure your Spring applications to use full AspectJ via LTW (load-time weaving) instead.
As you noticed, my sample code does not use Spring at all because AspectJ is completely independent of Spring and works with any Java application (or other JVM languages, too).
Refactoring idea
Now what should you do in order to get rid of passing around tracing information (Transaction objects), polluting your core application code and tangling it with trace calls?
You extract transaction tracing into an aspect taking care of all trace(..) and stop() calls.
Unfortunately your Transaction class contains different types of information and does different things, so you cannot completely get rid of context information about how to trace for each affected method. But at least you can extract that context information from the method bodies and transform it into a declarative form using annotations with parameters.
These annotations can be targeted by an aspect taking care of handling transaction tracing.
Added and updated code, iteration 1
Annotations related to transaction tracing:
package de.scrum_master.trace;
import static java.lang.annotation.ElementType.METHOD;
import static java.lang.annotation.RetentionPolicy.RUNTIME;
import java.lang.annotation.Retention;
import java.lang.annotation.Target;
#Retention(RUNTIME)
#Target(METHOD)
public #interface TransactionEntryPoint {}
package de.scrum_master.trace;
import static java.lang.annotation.ElementType.METHOD;
import static java.lang.annotation.RetentionPolicy.RUNTIME;
import java.lang.annotation.Retention;
import java.lang.annotation.Target;
#Retention(RUNTIME)
#Target(METHOD)
public #interface TransactionTrace {
String message() default "__METHOD_NAME__";
TraceCategory category() default TraceCategory.Code;
String addStat() default "";
String incrementStat() default "";
}
Refactored application classes with annotations:
package de.scrum_master.app;
import java.util.ArrayList;
import java.util.List;
import de.scrum_master.trace.TransactionTrace;
public class Order {
private int id;
public Order(int id) {
this.id = id;
}
#TransactionTrace(message = "", addStat = "number of items")
public List<Item> getItems() {
List<Item> items = new ArrayList<>();
int offset = id == 12345 ? 0 : 1;
items.add(new Item(11 + offset));
items.add(new Item(22 + offset));
items.add(new Item(33 + offset));
return items;
}
}
Nothing much here, only added an annotation to getItems(). But the sample application class changes massively, getting much cleaner and simpler:
package de.scrum_master.app;
import de.scrum_master.trace.TraceCategory;
import de.scrum_master.trace.TransactionEntryPoint;
import de.scrum_master.trace.TransactionTrace;
public class Example {
private OfferRepository offerRepository = new OfferRepository();
#TransactionEntryPoint
#TransactionTrace
public Order getOrderEntryPoint(int orderId) {
sleep(100);
Order order = getOrder(orderId);
return order;
}
#TransactionTrace
private Order getOrder(int orderId) {
sleep(200);
Order order = new Order(orderId);
for (Item item : order.getItems()) {
SpecialOffer offer = getSpecialOffer(item);
// Do something with special offers
}
return order;
}
#TransactionTrace(category = TraceCategory.Database, incrementStat = "specialOffers")
private SpecialOffer getSpecialOffer(Item item) {
sleep(50);
SpecialOffer specialOffer = offerRepository.getByItem(item);
return specialOffer;
}
private void sleep(long millis) {
try {
Thread.sleep(millis);
} catch (InterruptedException e) {
e.printStackTrace();
}
}
public static void main(String[] args) {
new Example().getOrderEntryPoint(12345);
new Example().getOrderEntryPoint(23456);
}
}
See? Except for a few annotations there is nothing left of the transaction tracing logic, the application code only takes care of its core concern. If you also remove the sleep() method which only makes the application slower for demonstration purposes (because we want some nice statistics with measured times >0 ms), the class gets even more compact.
But of course we need to put the transaction tracing logic somewhere, more precisely modularise it into an AspectJ aspect:
Transaction tracing aspect:
package de.scrum_master.trace;
import java.lang.reflect.Array;
import java.util.Arrays;
import java.util.Collection;
import java.util.stream.Collectors;
import org.aspectj.lang.JoinPoint;
import org.aspectj.lang.ProceedingJoinPoint;
import org.aspectj.lang.annotation.After;
import org.aspectj.lang.annotation.Around;
import org.aspectj.lang.annotation.Aspect;
import org.aspectj.lang.annotation.Pointcut;
import org.aspectj.lang.reflect.MethodSignature;
#Aspect("percflow(entryPoint())")
public class TransactionTraceAspect {
private static TransactionManager transactionManager = new TransactionManager();
private Transaction transaction = transactionManager.create();
#Pointcut("execution(* *(..)) && #annotation(de.scrum_master.trace.TransactionEntryPoint)")
private static void entryPoint() {}
#Around("execution(* *(..)) && #annotation(transactionTrace)")
public Object doTrace(ProceedingJoinPoint joinPoint, TransactionTrace transactionTrace) throws Throwable {
preTrace(transactionTrace, joinPoint);
Object result = joinPoint.proceed();
postTrace(transactionTrace);
addStat(transactionTrace, result);
incrementStat(transactionTrace, result);
return result;
}
private void preTrace(TransactionTrace transactionTrace, ProceedingJoinPoint joinPoint) {
String traceMessage = transactionTrace.message();
if ("".equals(traceMessage))
return;
MethodSignature signature = (MethodSignature) joinPoint.getSignature();
if ("__METHOD_NAME__".equals(traceMessage)) {
traceMessage = signature.getName() + "(";
traceMessage += Arrays.stream(joinPoint.getArgs()).map(arg -> arg.toString()).collect(Collectors.joining(", "));
traceMessage += ")";
}
transaction.trace(traceMessage, transactionTrace.category());
}
private void postTrace(TransactionTrace transactionTrace) {
if ("".equals(transactionTrace.message()))
return;
transaction.stop();
}
private void addStat(TransactionTrace transactionTrace, Object result) {
if ("".equals(transactionTrace.addStat()) || result == null)
return;
if (result instanceof Collection)
transaction.addStat(transactionTrace.addStat(), ((Collection<?>) result).size());
else if (result.getClass().isArray())
transaction.addStat(transactionTrace.addStat(), Array.getLength(result));
}
private void incrementStat(TransactionTrace transactionTrace, Object result) {
if ("".equals(transactionTrace.incrementStat()) || result == null)
return;
transaction.incrementStat(transactionTrace.incrementStat(), 1);
}
#After("entryPoint()")
public void logFinishedTransaction(JoinPoint joinPoint) {
System.out.println(transaction);
}
}
Let me explain what this aspect does:
#Pointcut(..) entryPoint() says: Find me all methods in the code annotated by #TransactionEntryPoint. This pointcut is used in two places:
#Aspect("percflow(entryPoint())") says: Create one aspect instance for each control flow beginning at a transaction entry point.
#After("entryPoint()") logFinishedTransaction(..) says: Execute this advice (AOP terminology for a method linked to a pointcut) after an entry point methods is finished. The corresponding method just prints the transaction statistics just like in the original code at the end of Example.getOrderEntryPoint(..).
#Around("execution(* *(..)) && #annotation(transactionTrace)") doTrace(..)says: Wrap methods annotated by TransactionTrace and do the following (method body):
add new trace element and start measuring time
execute original (wrapped) method and store result
update trace element with measured time
add one type of statistics (optional)
increment another type of statistics (optional)
return wrapped method's result to its caller
The private methods are just helpers for the #Around advice.
The console log when running the updated Example class and active AspectJ is:
Transaction {
uuid = 4529d325-c604-441d-8997-45ca659abb14
stats = {
specialOffers=2
number of items=3
}
events = {
TraceEvent[name=getOrderEntryPoint(12345), category=Code, durationInMs=468, finished=true]
TraceEvent[name=getOrder(12345), category=Code, durationInMs=366, finished=true]
TraceEvent[name=getSpecialOffer(Item[id=11]), category=Database, durationInMs=59, finished=true]
TraceEvent[name=getSpecialOffer(Item[id=22]), category=Database, durationInMs=50, finished=true]
TraceEvent[name=getSpecialOffer(Item[id=33]), category=Database, durationInMs=51, finished=true]
}
}
Transaction {
uuid = ef76a996-8621-478b-a376-e9f7a729a501
stats = {
specialOffers=2
number of items=3
}
events = {
TraceEvent[name=getOrderEntryPoint(23456), category=Code, durationInMs=452, finished=true]
TraceEvent[name=getOrder(23456), category=Code, durationInMs=351, finished=true]
TraceEvent[name=getSpecialOffer(Item[id=12]), category=Database, durationInMs=50, finished=true]
TraceEvent[name=getSpecialOffer(Item[id=23]), category=Database, durationInMs=50, finished=true]
TraceEvent[name=getSpecialOffer(Item[id=34]), category=Database, durationInMs=50, finished=true]
}
}
You see, it looks almost identical to the original application.
Idea for further simplification, iteration 2
When reading method Example.getOrder(int orderId) I was wondering why you are calling order.getItems(), looping over it and calling getSpecialOffer(item) inside the loop. In your sample code you do not use the results for anything other than updating the transaction trace object. I am assuming that in your real code you do something with the order and with the special offers in that method.
But just in case you really do not need those calls inside that method, I suggest
you factor the calls out right into the aspect, getting rid of the TransactionTrace annotation parameters String addStat() and String incrementStat().
The Example code would get even simpler and
the annotation #TransactionTrace(message = "", addStat = "number of items") in class would go away, too.
I am leaving this refactoring to you if you think it makes sense.

Customized parameter logging when using aspect oriented programing

All the examples I've seen that use aspect oriented programming for logging either log just class, method name and duration, and if they log parameters and return values they simply use ToString(). I need to have more control over what is logged. For example I want to skip passwords, or in some cases log all properties of an object but in other cases just the id property.
Any suggestions? I looked at AspectJ in Java and Unity interception in C# and could not find a solution.
You could try introducing parameter annotations to augment your parameters with some attributes. One of those attributes could signal to skip logging the parameter, another one could be used to specify a converter class for the string representation.
With the following annotations:
#Documented
#Retention(RetentionPolicy.RUNTIME)
#Target(ElementType.METHOD)
public #interface Log {
}
#Documented
#Retention(RetentionPolicy.RUNTIME)
#Target(ElementType.PARAMETER)
public #interface SkipLogging {
}
#Documented
#Retention(RetentionPolicy.RUNTIME)
#Target(ElementType.PARAMETER)
public #interface ToStringWith {
Class<? extends Function<?, String>> value();
}
the aspect could look like this:
import java.lang.reflect.Parameter;
import java.util.function.Function;
import java.util.stream.Collectors;
import java.util.stream.IntStream;
import org.aspectj.lang.reflect.MethodSignature;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
public aspect LoggingAspect {
private final static Logger logger = LoggerFactory.getLogger(LoggingAspect.class);
pointcut loggableMethod(): execution(#Log * *..*.*(..));
before(): loggableMethod() {
MethodSignature signature = (MethodSignature) thisJoinPoint.getSignature();
Parameter[] parameters = signature.getMethod()
.getParameters();
String message = IntStream.range(0, parameters.length)
.filter(i -> this.isLoggable(parameters[i]))
.<String>mapToObj(i -> toString(parameters[i], thisJoinPoint.getArgs()[i]))
.collect(Collectors.joining(", ",
"method execution " + signature.getName() + "(", ")"));
Logger methodLogger = LoggerFactory.getLogger(
thisJoinPointStaticPart.getSignature().getDeclaringType());
methodLogger.debug(message);
}
private boolean isLoggable(Parameter parameter) {
return parameter.getAnnotation(SkipLogging.class) == null;
}
private String toString(Parameter parameter, Object value) {
ToStringWith toStringWith = parameter.getAnnotation(ToStringWith.class);
if (toStringWith != null) {
Class<? extends Function<?, String>> converterClass =
toStringWith.value();
try {
#SuppressWarnings("unchecked")
Function<Object, String> converter = (Function<Object, String>)
converterClass.newInstance();
String str = converter.apply(value);
return String.format("%s='%s'", parameter.getName(), str);
} catch (Exception e) {
logger.error("Couldn't instantiate toString converter for logging "
+ converterClass.getName(), e);
return String.format("%s=<error converting to string>",
parameter.getName());
}
} else {
return String.format("%s='%s'", parameter.getName(), String.valueOf(value));
}
}
}
Test code:
public static class SomethingToStringConverter implements Function<Something, String> {
#Override
public String apply(Something something) {
return "Something nice";
}
}
#Log
public void test(
#ToStringWith(SomethingToStringConverter.class) Something something,
String string,
#SkipLogging Class<?> cls,
Object object) {
}
public static void main(String[] args) {
// execution of this method should log the following message:
// method execution test(something='Something nice', string='some string', object='null')
test(new Something(), "some string", Object.class, null);
}
I used Java 8 Streams API in my answer for it's compactness, you could convert the code to normal Java code if you don't use Java 8 features or need better efficiency. It's just to give you an idea.

Arquillian Graphene #Location placeholder

I'm learning Arquillian right now I wonder how to create page that has a placeholder inside the path. For example:
#Location("/posts/{id}")
public class BlogPostPage {
public String getContent() {
// ...
}
}
or
#Location("/posts/{name}")
#Location("/specific-page?requiredParam={value}")
I have looking for an answer on graphine and arquillian reference guides without success. I used library from other language that have support for page-objects, but it has build-in support for placeholders.
AFAIK there is nothing like this implemented in Graphene.
To be honest, I'm not sure how this should behave - how would you pass the values...?
Apart from that, I think that it could be also limited by Java annotation abilities https://stackoverflow.com/a/10636320/6835063
This is not possible currently in Graphene. I've created ARQGRA-500.
It's possible to extend Graphene to add dynamic parameters now. Here's how. (Arquillian 1.1.10.Final, Graphene 2.1.0.Final.)
Create an interface.
import java.util.Map;
public interface LocationParameterProvider {
Map<String, String> provideLocationParameters();
}
Create a custom LocationDecider to replace the corresponding Graphene's one. I replace the HTTP one. This Decider will add location parameters to the URI, if it sees that the test object implements our interface.
import java.io.UnsupportedEncodingException;
import java.net.URLEncoder;
import java.util.Map;
import java.util.Map.Entry;
import org.jboss.arquillian.core.api.Instance;
import org.jboss.arquillian.core.api.annotation.Inject;
import org.jboss.arquillian.graphene.location.decider.HTTPLocationDecider;
import org.jboss.arquillian.graphene.spi.location.Scheme;
import org.jboss.arquillian.test.spi.context.TestContext;
public class HTTPParameterizedLocationDecider extends HTTPLocationDecider {
#Inject
private Instance<TestContext> testContext;
#Override
public Scheme canDecide() {
return new Scheme.HTTP();
}
#Override
public String decide(String location) {
String uri = super.decide(location);
// not sure, how reliable this method of getting the current test object is
// if it breaks, there is always a possibility of observing
// org.jboss.arquillian.test.spi.event.suite.TestLifecycleEvent's (or rather its
// descendants) and storing the test object in a ThreadLocal
Object testObject = testContext.get().getActiveId();
if (testObject instanceof LocationParameterProvider) {
Map<String, String> locationParameters =
((LocationParameterProvider) testObject).provideLocationParameters();
StringBuilder uriParams = new StringBuilder(64);
boolean first = true;
for (Entry<String, String> param : locationParameters.entrySet()) {
uriParams.append(first ? '?' : '&');
first = false;
try {
uriParams.append(URLEncoder.encode(param.getKey(), "UTF-8"));
uriParams.append('=');
uriParams.append(URLEncoder.encode(param.getValue(), "UTF-8"));
} catch (UnsupportedEncodingException e) {
throw new RuntimeException(e);
}
}
uri += uriParams.toString();
}
return uri;
}
}
Our LocationDecider must be registered to override the Graphene's one.
import org.jboss.arquillian.core.spi.LoadableExtension;
import org.jboss.arquillian.graphene.location.decider.HTTPLocationDecider;
import org.jboss.arquillian.graphene.spi.location.LocationDecider;
public class MyArquillianExtension implements LoadableExtension {
#Override
public void register(ExtensionBuilder builder) {
builder.override(LocationDecider.class, HTTPLocationDecider.class,
HTTPParameterizedLocationDecider.class);
}
}
MyArquillianExtension should be registered via SPI, so create a necessary file in your test resources, e.g. for me the file path is src/test/resources/META-INF/services/org.jboss.arquillian.core.spi.LoadableExtension. The file must contain a fully qualified class name of MyArquillianExtension.
And that's it. Now you can provide location parameters in a test.
import java.util.HashMap;
import java.util.Map;
import org.jboss.arquillian.graphene.page.InitialPage;
import org.jboss.arquillian.graphene.page.Location;
import org.junit.Test;
public class TestyTest implements LocationParameterProvider {
#Override
public Map<String, String> provideLocationParameters() {
Map<String, String> params = new HashMap<>();
params.put("mykey", "myvalue");
return params;
}
#Test
public void test(#InitialPage TestPage page) {
}
#Location("MyTestView.xhtml")
public static class TestPage {
}
}
I've focused on parameters specifically, but hopefully this paves the way for other dynamic path manipulations.
Of course this doesn't fix the Graphene.goTo API. This means before using goTo you have to provide parameters via this roundabout provideLocationParameters way. It's weird. You can make your own alternative API, goTo that accepts parameters, and modify your LocationDecider to support other ParameterProviders.

AspectJ, Intertype Definition

I am receiving this error when I compile
The type XXX must implement the inherited abstract method
I have three files
A default implementation [com.SafeReaderIMPL.java]
public class SafeReaderIMPL implements ISafeReader {
private boolean successfulRead;
public SafeReaderIMPL() {
successfulRead = true;
}
protected void fail() {
successfulRead = false;
}
#Override
public boolean isSuccessfulRead() {
return successfulRead;
}
}
An interface file [com.ISafeReader.java]
public interface ISafeReader {
public boolean isSuccessfulRead();
}
An apsect (using annotations) [com.SafeReaderAspect.java]
#Aspect
public class SafeReaderAspect {
#DeclareParents(value = "com.BadReader", defaultImpl = SafeReaderIMPL.class)
public ISafeReader implementedInterface;
#AfterThrowing(pointcut = "execution(* *.*(..)) && this(m)", throwing = "e")
public void handleBadRead(JoinPoint joinPoint, ISafeReader m, Throwable e) {
((SafeReaderIMPL)m).fail();
}
}
And a Test Class [com.BadReader]
public class BadReader {
public void fail() throws Throwable {
throw new Throwable();
}
}
I compile the first three files in a separate jar using
ajc -source 1.8 -sourceroots . -outjar aspectLib.jar
I then compile the second file using the aspectLib like so
ajc -source 1.8 -sourceroots . -aspectpath ./aspectLib.jar -outjar common.jar
When I go to compile the second jar I get the error. I am using the latest stable version of AspectJ 1.8.3
BadReader.java:10 [error] The type BadReader must implement the
inherited abstract method ISafeReader.isSuccessfulRead() public class
BadReader {
^^^^^^^^
The problem is not two-step compilation as such, but the fact that #DeclareParents in #AspectJ syntax in not 100% compatible with declare parents in native syntax. Actually, #DeclareParents for introducing default interface implementations is superseded by #DeclareMixin (see this bug ticket), but the downside of the mixin approach is that you do not have a real A implements B scenario there, i.e. you cannot cast as you wish in your after-throwing advice, so this is also not a good option in your case.
So what do you do if you want to keep the two-step compilation approach? Just use native syntax:
Interface:
package com;
public interface ISafeReader {
boolean isSuccessfulRead();
}
Default implementation:
package com;
public class SafeReaderIMPL implements ISafeReader {
private boolean successfulRead;
public SafeReaderIMPL() { successfulRead = true; }
public void fail() { successfulRead = false; }
#Override public boolean isSuccessfulRead() { return successfulRead; }
}
ITD aspect:
package com;
public aspect SafeReaderAspect {
declare parents : com.BadReader implements SafeReaderIMPL;
after(ISafeReader safeReader) throwing : execution(* *(..)) && this(safeReader) {
System.out.println(thisJoinPoint + " - calling 'fail()' before rethrowing error");
((SafeReaderIMPL) safeReader).fail();
}
}
ITD target class with sample main method:
package com;
public class BadReader {
public void doSomething() {
throw new RuntimeException("my error");
}
public static void main(String[] args) {
BadReader badReader = new BadReader();
System.out.println("badReader.isSuccessfulRead() = " + badReader.isSuccessfulRead());
try { badReader.doSomething(); }
catch(Throwable t) { System.out.println(t); }
System.out.println("badReader.isSuccessfulRead() = " + badReader.isSuccessfulRead());
}
}
Now you can use the two-stage compilation approach.
Console output:
badReader.isSuccessfulRead() = true
execution(void com.BadReader.doSomething()) - calling 'fail()' before rethrowing error
java.lang.RuntimeException: my error
badReader.isSuccessfulRead() = false
The problem is due to the two-step compilation. During the second step, ajc needs the source code of SafeReaderIMPL to be able to weave BadReader, but it cannot find it into aspectLib.jar
In fact, if you try compiling in a single step (I did), it compiles and runs.
Unfortunately I don't know a way to fix this without providing the source code during the second compile step, which I suppose would render the whole two-step approach a bit pointless.

Gradle : how to use BuildConfig in an android-library with a flag that gets set in an app

My (gradle 1.10 and gradle plugin 0.8)-based android project consists of a big android-library that is a dependency for 3 different android-apps
In my library, I would love to be able to use a structure like this
if (BuildConfig.SOME_FLAG) {
callToBigLibraries()
}
as proguard would be able to reduce the size of the produced apk, based on the final value of SOME_FLAG
But I can't figure how to do it with gradle as :
* the BuildConfig produced by the library doesn't have the same package name than the app
* I have to import the BuildConfig with the library package in the library
* The apk of an apps includes the BuildConfig with the package of the app but not the one with the package of the library.
I tried without success to play with BuildTypes and stuff like
release {
// packageNameSuffix "library"
buildConfigField "boolean", "SOME_FLAG", "true"
}
debug {
//packageNameSuffix "library"
buildConfigField "boolean", "SOME_FLAG", "true"
}
What is the right way to builds a shared BuildConfig for my library and my apps whose flags will be overridden at build in the apps?
As a workaround, you can use this method, which uses reflection to get the field value from the app (not the library):
/**
* Gets a field from the project's BuildConfig. This is useful when, for example, flavors
* are used at the project level to set custom fields.
* #param context Used to find the correct file
* #param fieldName The name of the field-to-access
* #return The value of the field, or {#code null} if the field is not found.
*/
public static Object getBuildConfigValue(Context context, String fieldName) {
try {
Class<?> clazz = Class.forName(context.getPackageName() + ".BuildConfig");
Field field = clazz.getField(fieldName);
return field.get(null);
} catch (ClassNotFoundException e) {
e.printStackTrace();
} catch (NoSuchFieldException e) {
e.printStackTrace();
} catch (IllegalAccessException e) {
e.printStackTrace();
}
return null;
}
To get the DEBUG field, for example, just call this from your Activity:
boolean debug = (Boolean) getBuildConfigValue(this, "DEBUG");
I have also shared this solution on the AOSP Issue Tracker.
Update: With newer versions of the Android Gradle plugin publishNonDefault is deprecated and has no effect anymore. All variants are now published.
The following solution/workaround works for me. It was posted by some guy in the google issue tracker:
Try setting publishNonDefault to true in the library project:
android {
...
publishNonDefault true
...
}
And add the following dependencies to the app project that is using the library:
dependencies {
releaseCompile project(path: ':library', configuration: 'release')
debugCompile project(path: ':library', configuration: 'debug')
}
This way, the project that uses the library includes the correct build type of the library.
You can't do what you want, because BuildConfig.SOME_FLAG isn't going to get propagated properly to your library; build types themselves aren't propagated to libraries -- they're always built as RELEASE. This is bug https://code.google.com/p/android/issues/detail?id=52962
To work around it: if you have control over all of the library modules, you could make sure that all the code touched by callToBigLibraries() is in classes and packages that you can cleave off cleanly with ProGuard, then use reflection so that you can access them if they exist and degrade gracefully if they don't. You're essentially doing the same thing, but you're making the check at runtime instead of compile time, and it's a little harder.
Let me know if you're having trouble figuring out how to do this; I could provide a sample if you need it.
I use a static BuildConfigHelper class in both the app and the library, so that I can have the packages BuildConfig set as final static variables in my library.
In the application, place a class like this:
package com.yourbase;
import com.your.application.BuildConfig;
public final class BuildConfigHelper {
public static final boolean DEBUG = BuildConfig.DEBUG;
public static final String APPLICATION_ID = BuildConfig.APPLICATION_ID;
public static final String BUILD_TYPE = BuildConfig.BUILD_TYPE;
public static final String FLAVOR = BuildConfig.FLAVOR;
public static final int VERSION_CODE = BuildConfig.VERSION_CODE;
public static final String VERSION_NAME = BuildConfig.VERSION_NAME;
}
And in the library:
package com.your.library;
import android.support.annotation.Nullable;
import java.lang.reflect.Field;
public class BuildConfigHelper {
private static final String BUILD_CONFIG = "com.yourbase.BuildConfigHelper";
public static final boolean DEBUG = getDebug();
public static final String APPLICATION_ID = (String) getBuildConfigValue("APPLICATION_ID");
public static final String BUILD_TYPE = (String) getBuildConfigValue("BUILD_TYPE");
public static final String FLAVOR = (String) getBuildConfigValue("FLAVOR");
public static final int VERSION_CODE = getVersionCode();
public static final String VERSION_NAME = (String) getBuildConfigValue("VERSION_NAME");
private static boolean getDebug() {
Object o = getBuildConfigValue("DEBUG");
if (o != null && o instanceof Boolean) {
return (Boolean) o;
} else {
return false;
}
}
private static int getVersionCode() {
Object o = getBuildConfigValue("VERSION_CODE");
if (o != null && o instanceof Integer) {
return (Integer) o;
} else {
return Integer.MIN_VALUE;
}
}
#Nullable
private static Object getBuildConfigValue(String fieldName) {
try {
Class c = Class.forName(BUILD_CONFIG);
Field f = c.getDeclaredField(fieldName);
f.setAccessible(true);
return f.get(null);
} catch (Exception e) {
e.printStackTrace();
return null;
}
}
}
Then, anywhere in your library where you want to check BuildConfig.DEBUG, you can check BuildConfigHelper.DEBUG and access it from anywhere without a context, and the same for the other properties. I did it this way so that the library will work with all my applications, without needing to pass a context in or set the package name some other way, and the application class only needs the import line changed to suit when adding it into a new application
Edit: I'd just like to reiterate, that this is the easiest (and only one listed here) way to get the values to be assigned to final static variables in the library from all of your applications without needing a context or hard coding the package name somewhere, which is almost as good as having the values in the default library BuildConfig anyway, for the minimal upkeep of changing that import line in each application.
For the case where the applicationId is not the same as the package (i.e. multiple applicationIds per project) AND you want to access from a library project:
Use Gradle to store the base package in resources.
In main/AndroidManifest.xml:
android {
applicationId "com.company.myappbase"
// note: using ${applicationId} here will be exactly as above
// and so NOT necessarily the applicationId of the generated APK
resValue "string", "build_config_package", "${applicationId}"
}
In Java:
public static boolean getDebug(Context context) {
Object obj = getBuildConfigValue("DEBUG", context);
if (obj instanceof Boolean) {
return (Boolean) o;
} else {
return false;
}
}
private static Object getBuildConfigValue(String fieldName, Context context) {
int resId = context.getResources().getIdentifier("build_config_package", "string", context.getPackageName());
// try/catch blah blah
Class<?> clazz = Class.forName(context.getString(resId) + ".BuildConfig");
Field field = clazz.getField(fieldName);
return field.get(null);
}
use both
my build.gradle
// ...
productFlavors {
internal {
// applicationId "com.elevensein.sein.internal"
applicationIdSuffix ".internal"
resValue "string", "build_config_package", "com.elevensein.sein"
}
production {
applicationId "com.elevensein.sein"
}
}
I want to call like below
Boolean isDebug = (Boolean) BuildConfigUtils.getBuildConfigValue(context, "DEBUG");
BuildConfigUtils.java
public class BuildConfigUtils
{
public static Object getBuildConfigValue (Context context, String fieldName)
{
Class<?> buildConfigClass = resolveBuildConfigClass(context);
return getStaticFieldValue(buildConfigClass, fieldName);
}
public static Class<?> resolveBuildConfigClass (Context context)
{
int resId = context.getResources().getIdentifier("build_config_package",
"string",
context.getPackageName());
if (resId != 0)
{
// defined in build.gradle
return loadClass(context.getString(resId) + ".BuildConfig");
}
// not defined in build.gradle
// try packageName + ".BuildConfig"
return loadClass(context.getPackageName() + ".BuildConfig");
}
private static Class<?> loadClass (String className)
{
Log.i("BuildConfigUtils", "try class load : " + className);
try {
return Class.forName(className);
} catch (ClassNotFoundException e) {
e.printStackTrace();
}
return null;
}
private static Object getStaticFieldValue (Class<?> clazz, String fieldName)
{
try { return clazz.getField(fieldName).get(null); }
catch (NoSuchFieldException e) { e.printStackTrace(); }
catch (IllegalAccessException e) { e.printStackTrace(); }
return null;
}
}
For me this is the ONLY ONE AND ACCEPTABLE* SOLUTION TO determine the ANDROID APPLICATION BuildConfig.class:
// base entry point
// abstract application
// which defines the method to obtain the desired class
// the definition of the application is contained in the library
// that wants to access the method or in a superior library package
public abstract class BasApp extends android.app.Application {
/*
* GET BUILD CONFIG CLASS
*/
protected Class<?> getAppBuildConfigClass();
// HELPER METHOD TO CAST CONTEXT TO BASE APP
public static BaseApp getAs(android.content.Context context) {
BaseApp as = getAs(context, BaseApp.class);
return as;
}
// HELPER METHOD TO CAST CONTEXT TO SPECIFIC BASEpp INHERITED CLASS TYPE
public static <I extends BaseApp> I getAs(android.content.Context context, Class<I> forCLass) {
android.content.Context applicationContext = context != null ?context.getApplicationContext() : null;
return applicationContext != null && forCLass != null && forCLass.isAssignableFrom(applicationContext.getClass())
? (I) applicationContext
: null;
}
// STATIC HELPER TO GET BUILD CONFIG CLASS
public static Class<?> getAppBuildConfigClass(android.content.Context context) {
BaseApp as = getAs(context);
Class buildConfigClass = as != null
? as.getAppBuildConfigClass()
: null;
return buildConfigClass;
}
}
// FINAL APP WITH IMPLEMENTATION
// POINTING TO DESIRED CLASS
public class MyApp extends BaseApp {
#Override
protected Class<?> getAppBuildConfigClass() {
return somefinal.app.package.BuildConfig.class;
}
}
USAGE IN LIBRARY:
Class<?> buildConfigClass = BaseApp.getAppBuildConfigClass(Context);
if(buildConfigClass !- null) {
// do your job
}
*there are couple of things need to be watched out:
getApplicationContext() - could return a context which is not an App ContexWrapper implementation - see what Applicatio class extends & get to know of the possibilities of context wrapping
the class returned by final app could be loaded by different class loaders than those who will use it - depends of loader implementation and some principals typical (chierarchy, visibility) for loaders
everything depends on the implemmentation of as in this case simple DELEGATION!!! - the solution could be more sophisticetaded - i wanted only to show here the usage of DELEGATION pattern :)
** why i downwoted all of reflection based patterns because they all have weak points and they all in some certain conditions will fail:
Class.forName(className); - because of not speciified loader
context.getPackageName() + ".BuildConfig"
a) context.getPackageName() - "by default - else see b)" returns not package defined in manifest but application id (somtimes they both are the same), see how the manifest package property is used and its flow - at the end apt tool will replace it with applicaton id (see ComponentName class for example what the pkg stands for there)
b) context.getPackageName() - will return what the implementaio wants to :P
*** what to change in my solution to make it more flawless
replace class with its name that will drop the problems wchich could appear when many classes loaded with different loaders accessing / or are used to obtain a final result involving class (get to know what describes the equality between two classes (for a compiler at runtime) - in short a class equality defines not a self class but a pair which is constituted by the loader and the class. (some home work - try load a inner class with different loader and access it by outer class loaded with different loader) - it would turns out that we will get illegal access error :) even the inner class is in the same package has all modificators allowing access to it outer class :) compiler/linker "VM" treats them as two not related classes...