Kotlin dynamically compile a class from source code at runtime - kotlin

Is it possible to compile and instantiate Kotlin class at runtime? I'm talking about something like that but using Kotlin API: How do I programmatically compile and instantiate a Java class?
As example:
I'm getting full class definition as String:
val example = "package example\n" +
"\n" +
"fun main(args: Array<String>) {\n" +
" println(\"Hello World\")\n" +
"}\n"
And then inserting it into some class.kt and running it so I'm getting "Hello World" printed in console at runtime.

You might want to look at Kotlin Scripting, see https://github.com/andrewoma/kotlin-script
Alternatively, you'll need to write your own eval(kotlin-code-string-here) method which will dump the text inside blah.kt file for example, compile it using an external Kotlin compiler into blah.class then dynamically load those classes into the runtime using the Java Classloader doing something like this:
MainClass.class.classLoader.loadClass("com.mypackage.MyClass")
This might be very slow and unreliable.
Another no so great option is to make use of Rhino and run JavaScript inside your Kotlin code. So once again, you'll have an eval(kotlin-code-string-here) method which will dump the content to a blah.kt file, then you would use a Kotlin2JS compiler to compile it to JavaScript and directly execute the JavaScript inside Kotlin using Rhino which is not great either.
Another option is to make use of Kotlin Scripting or an external Kotlin compiler (in both cases, the Kotlin compiler will have to start up) and doing something like this will also allow you to execute dynamically, albeit, only on Unix systems.
Runtime.getRuntime().exec(""" "kotlin code here" > blah.kts | sh""")
I'm not aware of a clean solution for this, Kotlin was not designed to be run like like PHP / JavaScript / Python which just interprets text dynamically, it has to compile to bytecode first before it can do anything on the JVM; so in each scenario, you will need to compile that code first in one way or another, whether to bytecode or to javascript and in both cases load it into you application using the Java Classloader or Rhino.

Please check this solution for dependencies, jar resources, etc. Code below isn't enough for successful execution.
However, to compile dynamic class you can do the following:
val classLoader = Thread.currentThread().contextClassLoader
val engineManager = ScriptEngineManager(classLoader)
setIdeaIoUseFallback() // hack to have ability to do this from IntelliJ Idea context
val ktsEngine: ScriptEngine = engineManager.getEngineByExtension("kts")
ktsEngine.eval("object MyClass { val number = 123 } ")
println(ktsEngine.eval("MyClass.number"))
Please note: there is code injection possible here. Please be careful and use dedicated process or dedicated ClassLoader for this.

KotlinScript can be used to compile Kotlin source code (e.g. to generate a jar file that can then be loaded).
Here's a Java project which demonstrates this (code would be cleaner in Kotlin):
https://github.com/alexoooo/sample-kotlin-compile/blob/main/src/main/java/io/github/alexoooo/sample/compile/KotlinCompilerFacade.java
Note that the code you provide would be generated as a nested class (inside the script).
Here is a Kotlin version:
#KotlinScript
object KotlinDynamicCompiler {
//-----------------------------------------------------------------------------------------------------------------
const val scriptClassName = "__"
const val classNamePrefix = "${scriptClassName}$"
private val baseClassType: KotlinType = KotlinType(KotlinDynamicCompiler::class.java.kotlin)
private val contextClass: KClass<*> = ScriptCompilationConfiguration::class.java.kotlin
//-----------------------------------------------------------------------------------------------------------------
fun compile(
kotlinCode: String, outputJarFile: Path, classpathLocations: List<Path>, classLoader: ClassLoader
): String? {
Files.createDirectories(outputJarFile.parent)
val scriptCompilationConfiguration = createCompilationConfigurationFromTemplate(
baseClassType, defaultJvmScriptingHostConfiguration, contextClass
) {
jvm {
val classloaderClasspath: List<File> = classpathFromClassloader(classLoader, false)!!
val classpathFiles = classloaderClasspath + classpathLocations.map { it.toFile() }
updateClasspath(classpathFiles)
}
hostConfiguration(ScriptingHostConfiguration (defaultJvmScriptingHostConfiguration) {
jvm {
compilationCache(
CompiledScriptJarsCache { _, _ ->
outputJarFile.toFile()
}
)
}
})
}
val scriptCompilerProxy = ScriptJvmCompilerIsolated(defaultJvmScriptingHostConfiguration)
val result = scriptCompilerProxy.compile(
kotlinCode.toScriptSource(KotlinCode.scriptClassName), scriptCompilationConfiguration)
val errors = result.reports.filter { it.severity == ScriptDiagnostic.Severity.ERROR }
return when {
errors.isEmpty() -> null
else -> errors.joinToString(" | ")
}
}
}

Related

Why is the kotlin-gradle-plugin failing to create a PSIFile from CodeInsightTestFixture.configureByText?

I created an IntelliJ plugin using the template https://github.com/JetBrains/intellij-platform-plugin-template. The template comes with a test that runs on an XML file. I want to create a similar test for a Kotlin file. Here's the template test file plus my added test (test2):
package org.jetbrains.plugins.template
import com.intellij.ide.highlighter.XmlFileType
import com.intellij.psi.xml.XmlFile
import com.intellij.testFramework.TestDataPath
import com.intellij.testFramework.fixtures.BasePlatformTestCase
import com.intellij.util.PsiErrorElementUtil
#TestDataPath("\$CONTENT_ROOT/src/test/testData")
class MyPluginTest : BasePlatformTestCase() {
fun testXMLFile() {
val psiFile = myFixture.configureByText(XmlFileType.INSTANCE, "<foo>bar</foo>")
val xmlFile = assertInstanceOf(psiFile, XmlFile::class.java)
assertFalse(PsiErrorElementUtil.hasErrors(project, xmlFile.virtualFile))
assertNotNull(xmlFile.rootTag)
xmlFile.rootTag?.let {
assertEquals("foo", it.name)
assertEquals("bar", it.value.text)
}
}
override fun getTestDataPath() = "src/test/testData/rename"
fun testRename() {
myFixture.testRename("foo.xml", "foo_after.xml", "a2")
}
// Here's my test
fun test2() {
val fileText: String = """
package com.loganmay.test
data class MyClass(val myString: String)
""".trimIndent()
val psiFile = myFixture.configureByText("a.kt", fileText)
val xmlFile = assertInstanceOf(psiFile, XmlFile::class.java)
}
}
Without changing the build.gradle file, that test fails with:
Expected instance of: com.intellij.psi.xml.XmlFile actual: com.intellij.psi.impl.source.PsiPlainTextFileImpl
I want it to parse the text as a PsiFile that's also a KtFile. From various sources, I've been led to believe that the fixture is parsing it as a plain text file because the test project doesn't have access to the Kotlin compiler. So, I added:
dependencies {
testImplementation("org.jetbrains.kotlin:kotlin-gradle-plugin:1.7.10")
}
to the build.gradle. Then, when I run the test, configureByText throws an exception with a big trace, the root exception of which is:
Caused by: java.lang.Throwable: 'filetype.archive.display.name' is not found in java.util.PropertyResourceBundle#4ecbb519(messages.CoreBundle)
... 53 more
org.jetbrains.plugins.template.MyPluginTest > test2 FAILED
com.intellij.diagnostic.PluginException at ComponentManagerImpl.kt:511
Caused by: java.util.MissingResourceException at Registry.java:164
Does anyone have any insight into what the issue is or know how to resolve it?
Notes:
I also tried importing the kotlin compiler and casting psiFile as KtFile, which produced the same error, an idea I got from here
This project has a test like this that may be working
This post and this post recommend adding the kotlin gradle plugin, which I did
This question seems similar
Yann Cebron replied on the jetbrains help forum with an answer for Java, which also worked for Kotlin.
The solution is to add a dependency to the IntelliJ gradle plugin. The template comes with these lines in the build.gradle:
intellij {
pluginName.set(properties("pluginName"))
version.set(properties("platformVersion"))
type.set(properties("platformType"))
// Plugin Dependencies. Uses `platformPlugins` property from the gradle.properties file.
plugins.set(properties("platformPlugins").split(',').map(String::trim).filter(String::isNotEmpty))
}
So, didn't need to do anything there. In my gradle.properties, I added
platformPlugins = com.intellij.java, org.jetbrains.kotlin
To my plugin.xml, I added:
<depends>com.intellij.modules.java</depends>
<depends>org.jetbrains.kotlin</depends>
I was able to remove
dependencies {
testImplementation("org.jetbrains.kotlin:kotlin-gradle-plugin:1.7.10")
}
from the build.gradle which I mentioned above.
Now, the test works for Java and Kotlin files.

How to run kotest which are not tagged by default?

In the kotest framework, there is a way to group tests with custom tags and you can run the particular group by selecting via Gradle parameter like gradle test -Dkotest.tags="TestGroupOne"
I have two test cases one is with a tag and another one is without a tag
object Linux : Tag()
class MyTests : StringSpec({
"without tag" {
"hello".length shouldBe 5
}
"with tag".config(tags = setOf(Linux)) {
"world" should startWith("wo2")
}
})
Now if I run gradle build it runs both tests, but I would like to run the tests which are not tagged by default. In the above example, the test without tag should run if there is no parameter passed in gradle
One way to achieve this behaviour is by adding a task in build.gradle.kts file
val test by tasks.getting(Test::class) {
systemProperties = System.getProperties()
.toList()
.associate { it.first.toString() to it.second }
if(!systemProperties.containsKey("kotest.tags"))
systemProperties["kotest.tags"] = "!Linux"
}
As you can see, when there is no parameter passed for -Dkotest.tags I'm manually adding the value !Linux to the systemProperties so that the build script will run tests which are not tagged by default.
Question: Is there any better way to achieve this?
I even tried adding systemProp.gradle.kotest.tags="!Linux" in gradle.properties file but there is no effect.
Your solution is not very robust in the sense that you depend on the concrete tag that is used. It seems that there is no easier solution for that, because the syntax for tag expressions does not allow to write something like "!any".
However, it is possible to write a Kotest extension for what you need that looks like this:
import io.kotest.core.TagExpression
import io.kotest.core.config.ProjectConfiguration
import io.kotest.core.extensions.ProjectExtension
import io.kotest.core.extensions.TestCaseExtension
import io.kotest.core.project.ProjectContext
import io.kotest.core.test.TestCase
import io.kotest.core.test.TestResult
import io.kotest.engine.tags.runtimeTags
object NoTagsExtension : TestCaseExtension, ProjectExtension {
private lateinit var config: ProjectConfiguration
override suspend fun interceptProject(context: ProjectContext, callback: suspend (ProjectContext) -> Unit) {
config = context.configuration
callback(context)
}
override suspend fun intercept(testCase: TestCase, execute: suspend (TestCase) -> TestResult): TestResult {
return if (config.runtimeTags().expression == TagExpression.Empty.expression) {
if (testCase.spec.tags().isEmpty() && testCase.config.tags.isEmpty()) {
execute(testCase)
} else TestResult.Ignored("currently running only tests without tags")
} else execute(testCase)
}
}
The first function interceptProject is just there to obtain the project configuration in order to determine the specified set of tags for the current test run.
The second function intercept is for each test-case. There we determine if any tags have been specified. If no tags were specified (i.e. we have an empty tag expression), we skip all test where any tag has been configured at the spec or test-case. Otherwise, we execute the test normally, and it will then possibly ignored by Kotlin's built-in mechanisms, depending on its tags.
The extension can be activated project-wide in the ProjectConfig:
class ProjectConfig : AbstractProjectConfig() {
override fun extensions(): List<Extension> = super.extensions() + NoTagsExtension
}
Now, with the extension in place, only tests without tag run by default, regardless of what tags you use in your project.

Is there any way to iterate all fields of a data class without using reflection?

I know an alternative of reflection which is using javassist, but using javassist is a little bit complex. And because of lambda or some other features in koltin, the javassist doesn't work well sometimes. So is there any other way to iterate all fields of a data class without using reflection.
There are two ways. The first is relatively easy, and is essentially what's mentioned in the comments: assuming you know how many fields there are, you can unpack it and throw that into a list, and iterate over those. Or alternatively use them directly:
data class Test(val x: String, val y: String) {
fun getData() : List<Any> = listOf(x, y)
}
data class Test(val x: String, val y: String)
...
val (x, y) = Test("x", "y")
// And optionally throw those in a list
Although iterating like this is a slight extra step, this is at least one way you can relatively easy unpack a data class.
If you don't know how many fields there are (or you don't want to refactor), you have two options:
The first is using reflection. But as you mentioned, you don't want this.
That leaves a second, somewhat more complicated preprocessing option: annotations. Note that this only works with data classes you control - beyond that, you're stuck with reflection or implementations from the library/framework coder.
Annotations can be used for several things. One of which is metadata, but also code generation. This is a somewhat complicated alternative, and requires an additional module in order to get compile order right. If it isn't compiled in the right order, you'll end up with unprocessed annotations, which kinda defeats the purpose.
I've also created a version you can use with Gradle, but that's at the end of the post and it's a shortcut to implementing it yourself.
Note that I have only tested this with a pure Kotlin project - I've personally had problems with annotations between Java and Kotlin (although that was with Lombok), so I do not guarantee this will work at compile time if called from Java. Also note that this is complex, but avoids runtime reflection.
Explanation
The main issue here is a certain memory concern. This will create a new list every time you call the method, which makes it very similar to the method used by enums.
Local testing over 10000 iterations also show a general consistency of ~200 milliseconds to execute my approach, versus roughly 600 for reflection. However, for one iteration, mine uses ~20 milliseconds, where as reflection uses between 400 and 500 milliseconds. On one run, reflection took 1500 (!) milliseconds, while my approach took 18 milliseconds.
See also Java Reflection: Why is it so slow?. This appears to affect Kotlin as well.
The memory impact of creating a new list every time it's called can be noticeable though, but it'll also be collected so it shouldn't be that big a problem.
For reference, the code used for benchmarking (this will make sense after the rest of the post):
#AutoUnpack data class ExampleDataClass(val x: String, val y: Int, var m: Boolean)
fun main(a: Array<String>) {
var mine = 0L
var reflect = 0L
// for(i in 0 until 10000) {
var start = System.currentTimeMillis()
val cls = ExampleDataClass("example", 42, false)
for (field in cls) {
println(field)
}
mine += System.currentTimeMillis() - start
start = System.currentTimeMillis()
for (prop in ExampleDataClass::class.memberProperties) {
println("${prop.name} = ${prop.get(cls)}")
}
reflect += System.currentTimeMillis() - start
// }
println(mine)
println(reflect)
}
Setting up from scratch
This bases itself around two modules: a consumer module, and a processor module. The processor HAS to be in a separate module. It needs to be compiled separately from the consumer for the annotations to work properly.
First of all, your consumer project needs the annotation processor:
apply plugin: 'kotlin-kapt'
Additionally, you need to add stub generation. It complains it's unused while compiling, but without it, the generator seems to break for me:
kapt {
generateStubs = true
}
Now that that's in order, create a new module for the unpacker. Add the Kotlin plugin if you didn't already. You do not need the annotation processor Gradle plugin in this project. That's only needed by the consumer. You do, however, need kotlinpoet:
implementation "com.squareup:kotlinpoet:1.2.0"
This is to simplify aspects of the code generation itself, which is the important part here.
Now, create the annotation:
#Retention(AnnotationRetention.SOURCE)
#Target(AnnotationTarget.CLASS)
annotation class AutoUnpack
This is pretty much all you need. The retention is set to source because it has no value at runtime, and it only targets compile time.
Next, there's the processor itself. This is somewhat complicated, so bear with me. For reference, this uses the javax.* packages for annotation processing. Android note: this might work assuming you can plug in a Java module on a compileOnly scope without getting the Android SDK restrictions. As I mentioned earlier, this is mainly for pure Kotlin; Android might work, but I haven't tested that.
Anyways, the generator:
Because I couldn't find a way to generate the method into the class without touching the rest (and because according to this, that isn't possible), I'm going with an extension function generation approach.
You'll need a class UnpackCodeGenerator : AbstractProcessor(). In there, you'll first need two lines of boilerplate:
override fun getSupportedAnnotationTypes(): MutableSet<String> = mutableSetOf(AutoUnpack::class.java.name)
override fun getSupportedSourceVersion(): SourceVersion = SourceVersion.latest()
Moving on, there's the processing. Override the process function:
override fun process(annotations: MutableSet<out TypeElement>, roundEnv: RoundEnvironment): Boolean {
// Find elements with the annotation
val annotatedElements = roundEnv.getElementsAnnotatedWith(AutoUnpack::class.java)
if(annotatedElements.isEmpty()) {
// Self-explanatory
return false;
}
// Iterate the elements
annotatedElements.forEach { element ->
// Grab the name and package
val name = element.simpleName.toString()
val pkg = processingEnv.elementUtils.getPackageOf(element).toString()
// Then generate the class
generateClass(name,
if (pkg == "unnamed package") "" else pkg, // This is a patch for an issue where classes in the root
// package return package as "unnamed package" rather than empty,
// which breaks syntax because "package unnamed package" isn't legal.
element)
}
// Return true for success
return true;
}
This just sets up some of the later framework. The real magic happens in the generateClass function:
private fun generateClass(className: String, pkg: String, element: Element){
val elements = element.enclosedElements
val classVariables = elements
.filter {
val name = if (it.simpleName.contains("\$delegate"))
it.simpleName.toString().substring(0, it.simpleName.indexOf("$"))
else it.simpleName.toString()
it.kind == ElementKind.FIELD // Find fields
&& Modifier.STATIC !in it.modifiers // that aren't static (thanks to sebaslogen for issue #1: https://github.com/LunarWatcher/KClassUnpacker/issues/1)
// Additionally, we have to ignore private fields. Extension functions can't access these, and accessing
// them is a bad idea anyway. Kotlin lets you expose get without exposing set. If you, by default, don't
// allow access to the getter, there's a high chance exposing it is a bad idea.
&& elements.any { getter -> getter.kind == ElementKind.METHOD // find methods
&& getter.simpleName.toString() ==
"get${name[0].toUpperCase().toString() + (if (name.length > 1) name.substring(1) else "")}" // that matches the getter name (by the standard convention)
&& Modifier.PUBLIC in getter.modifiers // that are marked public
}
} // Grab the variables
.map {
// Map the name now. Also supports later filtering
if (it.simpleName.endsWith("\$delegate")) {
// Support by lazy
it.simpleName.subSequence(0, it.simpleName.indexOf("$"))
} else it.simpleName
}
if (classVariables.isEmpty()) return; // Self-explanatory
val file = FileSpec.builder(pkg, className)
.addFunction(FunSpec.builder("iterator") // For automatic unpacking in a for loop
.receiver(element.asType().asTypeName().copy()) // Add it as an extension function of the class
.addStatement("return listOf(${classVariables.joinToString(", ")}).iterator()") // add the return statement. Create a list, push an iterator.
.addModifiers(KModifier.PUBLIC, KModifier.OPERATOR) // This needs to be public. Because it's an iterator, the function also needs the `operator` keyword
.build()
).build()
// Grab the generate directory.
val genDir = processingEnv.options["kapt.kotlin.generated"]!!
// Then write the file.
file.writeTo(File(genDir, "$pkg/${element.simpleName.replace("\\.kt".toRegex(), "")}Generated.kt"))
}
All of the relevant lines have comments explaining use, in case you're not familiar with what this does.
Finally, in order to get the processor to process, you need to register it. In the module for the generator, add a file called javax.annotation.processing.Processor under main/resources/META-INF/services. In there you write:
com.package.of.UnpackCodeGenerator
From here, you need to link it using compileOnly and kapt. If you added it as a module to your project, you can do:
kapt project(":ClassUnpacker")
compileOnly project(":ClassUnpacker")
Alternative source setup:
Like I mentioned earlier, I bundled this into a jar for convenience. It's under the same license as SO uses (CC-BY-SA 3.0), and it contains the exact same code as in the answer (although compiled into a single project).
If you want to use this one, just add the Jitpack repo:
repositories {
// Other repos here
maven { url 'https://jitpack.io' }
}
And hook it up with:
kapt 'com.github.LunarWatcher:KClassUnpacker:v1.0.1'
compileOnly "com.github.LunarWatcher:KClassUnpacker:v1.0.1"
Note that the version here may not be up to date: the up to date list of versions is available here. The code in the post still aims to reflect the repo, but versions aren't really important enough to edit every time.
Usage
Regardless of which way you ended up using to get the annotations, the usage is relatively easy:
#AutoUnpack data class ExampleDataClass(val x: String, val y: Int, var m: Boolean)
fun main(a: Array<String>) {
val cls = ExampleDataClass("example", 42, false)
for(field in cls) {
println(field)
}
}
This prints:
example
42
false
Now you have a reflection-less way of iterating fields.
Note that local testing has been done partially with IntelliJ, but IntelliJ doesn't seem to like me - I've had various failed builds where gradlew clean && gradlew build from a command line oddly works fine. I'm not sure whether this is a local problem, or if this is a general problem, but you might have some issues like this if you build from IntelliJ.
Also, you might get errors if the build fails. The IntelliJ linter builds on top of the build directory for some sources, so if the build fails and the file with the extension function isn't generated, that'll cause it to appear as an error. Building usually fixes this when I tested (with both modules and from Jitpack).
You'll also likely have to enable the annotation processor setting if you use Android Studio or IntelliJ.
here is another idea, that i came up with, but am not satisfied with...but it has some pros and cons:
pros:
adding/removing fields to/from the data class causes compiler errors at field-iteration sites
no boiler-plate code needed
cons:
won't work if default values are defined for arguments
declaration:
data class Memento(
val testType: TestTypeData,
val notes: String,
val examinationTime: MillisSinceEpoch?,
val administeredBy: String,
val signature: SignatureViewHolder.SignatureData,
val signerName: String,
val signerRole: SignerRole
) : Serializable
iterating through all fields (can use this directly at call sites, or apply the Visitor pattern, and use this in the accept method to call all the visit methods):
val iterateThroughAllMyFields: Memento = someValue
Memento(
testType = iterateThroughAllMyFields.testType.also { testType ->
// do something with testType
},
notes = iterateThroughAllMyFields.notes.also { notes ->
// do something with notes
},
examinationTime = iterateThroughAllMyFields.examinationTime.also { examinationTime ->
// do something with examinationTime
},
administeredBy = iterateThroughAllMyFields.administeredBy.also { administeredBy ->
// do something with administeredBy
},
signature = iterateThroughAllMyFields.signature.also { signature ->
// do something with signature
},
signerName = iterateThroughAllMyFields.signerName.also { signerName ->
// do something with signerName
},
signerRole = iterateThroughAllMyFields.signerRole.also { signerRole ->
// do something with signerRole
}
)

Aspectj doesn't work with kotlin

i want to use aspectj aop in kotlin,here is my code:
my annotation in annotation.lazy_list:
Kotlin:
package anotation
#Retention(AnnotationRetention.RUNTIME)
#Target(AnnotationTarget.FUNCTION)
annotation class lazy_list
my aspectj aop class:
#Aspect
class ActiveListAop{
#Pointcut("execution(#annotation.lazy_list * *(..))")
fun profile() {
}
#Before("profile()")
fun testModeOnly(joinPoint: JoinPoint) {
println("123")
}
}
my usage:
#lazy_list
fun all():List<T>{
return lazy_obj?.all() as List<T>
}
when i call all() function , no error,but wont't print "123", why?
EDIT 9-2021 - there is a nice updated plugin for android that works well as an alternate to my original 2018 answer below: https://github.com/Ibotta/gradle-aspectj-pipeline-plugin
For what it's worth, we needed aspectJ weaving in our android project but really wanted to move to kotlin so we had to solve this problem. So the solutions in this thread using spring or maven didn't work for us. This is the solution for android gradle projects however, this WILL break incremental compilation and therefor slow down your build times and/or break something eventually. This gets us by until I can re-think our architecture and phase out aspectJ or (hopefully) android starts supporting it.
There is confusion in some of the answers and comments to the OP that kapt solves this, but kapt lets you do compile time annotation processing, not weaving. That is, annotation processors let you generate code based on annotations but do not let you inject logic into existing code.
This builds on top of this blog on adding aspectJ to android: https://fernandocejas.com/2014/08/03/aspect-oriented-programming-in-android
Your kotlin classes get compiled into byte code, just into a different directory. So this solution using the same process to weave the java classes but runs it again on the kotlin class files
at the top of your App/build.gradle add:
buildscript {
ext.aspectjVersion = '1.9.1'
dependencies {
classpath "org.aspectj:aspectjtools:$aspectjVersion"
}
}
At the bottom of your App/build.gradle add:
android.applicationVariants.all { variant ->
// add the versionName & versionCode to the apk file name
variant.outputs.all { output ->
def newPath = outputFileName.replace(".apk", "-${variant.versionName}.${variant.versionCode}.apk")
outputFileName = new File(outputFileName, newPath)
def fullName = ""
output.name.tokenize('-').eachWithIndex { token, index ->
fullName = fullName + (index == 0 ? token : token.capitalize())
}
JavaCompile javaCompile = variant.javaCompiler
MessageHandler handler = new MessageHandler(true)
javaCompile.doLast {
String[] javaArgs = ["-showWeaveInfo",
"-1.8",
"-inpath", javaCompile.destinationDir.toString(),
"-aspectpath", javaCompile.classpath.asPath,
"-d", javaCompile.destinationDir.toString(),
"-classpath", javaCompile.classpath.asPath,
"-bootclasspath", project.android.bootClasspath.join(
File.pathSeparator)]
String[] kotlinArgs = ["-showWeaveInfo",
"-1.8",
"-inpath", project.buildDir.path + "/tmp/kotlin-classes/" + fullName,
"-aspectpath", javaCompile.classpath.asPath,
"-d", project.buildDir.path + "/tmp/kotlin-classes/" + fullName,
"-classpath", javaCompile.classpath.asPath,
"-bootclasspath", project.android.bootClasspath.join(
File.pathSeparator)]
new Main().run(javaArgs, handler)
new Main().run(kotlinArgs, handler)
def log = project.logger
for (IMessage message : handler.getMessages(null, true)) {
switch (message.getKind()) {
case IMessage.ABORT:
case IMessage.ERROR:
case IMessage.FAIL:
log.error message.message, message.thrown
break
case IMessage.WARNING:
case IMessage.INFO:
log.info message.message, message.thrown
break
case IMessage.DEBUG:
log.debug message.message, message.thrown
break
}
}
}
}
spring + kotlin + AOP work nice, just go to http://start.spring.io/ and generate a project with AOP support, you can see a piece of build.gradle here...
buildscript {
ext {
kotlinVersion = '1.2.30'
springBootVersion = '2.0.0.RELEASE'
}
repositories {
mavenCentral()
}
dependencies {
classpath("org.springframework.boot:spring-boot-gradle-plugin:${springBootVersion}")
classpath("org.jetbrains.kotlin:kotlin-gradle-plugin:${kotlinVersion}")
classpath("org.jetbrains.kotlin:kotlin-allopen:${kotlinVersion}")
}
}
apply plugin: 'kotlin'
apply plugin: 'kotlin-spring'
apply plugin: 'org.springframework.boot'
...
dependencies {
compile('org.springframework.boot:spring-boot-starter-aop')
...
}
plugin kotlin-spring makes all classes open to allow AOP
Then, just declare your aspect as follows
#Aspect
#Component
class MyAspect {
...
Important: annotate your aspect class with #Aspect and #Component annotations
Piece of cake! :)
For annotation process in Kotlin, you must enable and use KAPT. Without this being added via Gradle or Maven plugin, nothing is going to work for annotation processing in Kotlin code.
The Kotlin plugin supports annotation processors like Dagger or DBFlow. In order for them to work with Kotlin classes, apply the kotlin-kapt plugin.
See also:
Pushing the limits of Kotlin annotation processing
kapt: Annotation Processing for Kotlin
Better Annotation Processing: Supporting Stubs in kapt
You can use freefair gradle plugin
buildscript {
repositories {
maven {
url "https://plugins.gradle.org/m2/"
}
}
dependencies {
classpath "io.freefair.gradle:aspectj-plugin:5.2.1"
}
}
apply plugin: "io.freefair.aspectj.post-compile-weaving"
So I think I've got a good (but wordy) solution for Android. At the time of writing I'm using Gradle 6.7, Android plugin 4.1.0, and AspectJ tools 1.9.6.
The gist of the problem is that:
Java is compiled by task compileDebugJavaWithJavac
Kotlin is compiled by task compileDebugKotlin
Gradle can run either one of these tasks, or both of them, or none
compileDebugJavaWithJavac depends on compileDebugKotlin
weaving Kotlin usually requires Java classes.
If you look at these points closely, you'll see that you can't do weaving as part of compiling Kotlin, as Java classes can be missing at this point. If you do that, you'll get warnings such as:
WARN: incorrect classpath: C:\Users\user\StudioProjects\myapp\app\build\intermediates\javac\debug\classes
and errors such as
ERROR: can’t determine modifiers of missing type myapp.Foo.Bar
So the better approach would be to postpone weaving until Java classes are compiled. But as you would be modifying files not as a part of compilation task, you lose incremental builds... Besides, this postponed weaving is super hard to get right—remember, none of the compile tasks might be actually scheduled for running!
The real solution is to wrap weaving in a Transform, which will produce a Gradle task with its own inputs and outputs. This means that you will not be polluting the files of compile tasks, and those tasks, as well as this task, will be, so to say, UP-TO-DATE-able. This requires quite a bit of code, but it's rather sensible!
First, put this in your project build.gradle.kts:
buildscript {
dependencies {
classpath("org.aspectj:aspectjtools:1.9.6")
}
}
This is needed to run weaving from inside “inside” the buildscript. If you want to run weaving in a separate process, which is a good idea on Windows, you will need the path of this jar, which you can get by adding the following to your app build.gradle.kts:
val weaving: Configuration by configurations.creating
dependencies {
weaving("org.aspectj:aspectjtools:1.9.6")
}
Finally, put AspectJ runtime on the classpath (app build.gradle.kts, note that I only need weaving in debug builds):
dependencies {
debugImplementation("org.aspectj:aspectjrt:1.9.6")
}
Now, here's my setup. I have a local logging library, :cats, which containts aspects that I want to weave. Logging statements are only inside my project, and not anywhere else. Also, I only want to run these in debug builds. So here's the transformation that “weaves cats” into the app (app's build.gradle.kts):
class TransformCats : Transform() {
override fun getName(): String = TransformCats::class.simpleName!!
override fun getInputTypes() = setOf(QualifiedContent.DefaultContentType.CLASSES)
// only look for annotations in app classes
// transformation will consume these and put woven classes in the output dir
override fun getScopes() = mutableSetOf(QualifiedContent.Scope.PROJECT)
// ...but also have the rest on our class path
// these will not be touched by the transformation
override fun getReferencedScopes() = mutableSetOf(QualifiedContent.Scope.SUB_PROJECTS,
QualifiedContent.Scope.EXTERNAL_LIBRARIES)
override fun isIncremental() = false
// only run on debug builds
override fun applyToVariant(variant: VariantInfo) = variant.isDebuggable
override fun transform(invocation: TransformInvocation) {
if (!invocation.isIncremental) {
invocation.outputProvider.deleteAll()
}
val output = invocation.outputProvider.getContentLocation(name, outputTypes,
scopes, Format.DIRECTORY)
if (output.isDirectory) FileUtils.deleteDirectoryContents(output)
FileUtils.mkdirs(output)
val input = mutableListOf<File>()
val classPath = mutableListOf<File>()
val aspectPath = mutableListOf<File>()
invocation.inputs.forEach { source ->
source.directoryInputs.forEach { dir ->
input.add(dir.file)
classPath.add(dir.file)
}
source.jarInputs.forEach { jar ->
input.add(jar.file)
classPath.add(jar.file)
}
}
invocation.referencedInputs.forEach { source ->
source.directoryInputs.forEach { dir ->
classPath.add(dir.file)
}
source.jarInputs.forEach { jar ->
classPath.add(jar.file)
if (jar.name == ":cats") aspectPath.add(jar.file)
}
}
weave(classPath, aspectPath, input, output)
}
}
android.registerTransform(TransformCats())
And here's the weaving code mentioned above:
// ajc gets hold of some files such as R.jar, and on Windows it leads to errors such as:
// The process cannot access the file because it is being used by another process
// to avoid these, weave in a process, which `javaexec` will helpfully launch for us.
fun weave(classPath: Iterable<File>, aspectPath: Iterable<File>, input: Iterable<File>, output: File) {
val runInAProcess = OperatingSystem.current().isWindows
val bootClassPath = android.bootClasspath
println(if (runInAProcess) ":: weaving in a process..." else ":: weaving...")
println(":: boot class path: $bootClassPath")
println(":: class path: $classPath")
println(":: aspect path: $aspectPath")
println(":: input: $input")
println(":: output: $output")
val arguments = listOf("-showWeaveInfo",
"-1.8",
"-bootclasspath", bootClassPath.asArgument,
"-classpath", classPath.asArgument,
"-aspectpath", aspectPath.asArgument,
"-inpath", input.asArgument,
"-d", output.absolutePath)
if (runInAProcess) {
javaexec {
classpath = weaving
main = "org.aspectj.tools.ajc.Main"
args = arguments
}
} else {
val handler = MessageHandler(true)
Main().run(arguments.toTypedArray(), handler)
val log = project.logger
for (message in handler.getMessages(null, true)) {
when (message.kind) {
IMessage.DEBUG -> log.debug("DEBUG " + message.message, message.thrown)
IMessage.INFO -> log.info("INFO: " + message.message, message.thrown)
IMessage.WARNING -> log.warn("WARN: " + message.message, message.thrown)
IMessage.FAIL,
IMessage.ERROR,
IMessage.ABORT -> log.error("ERROR: " + message.message, message.thrown)
}
}
}
}
val Iterable<File>.asArgument get() = joinToString(File.pathSeparator)
(The Windows part is using weaving configuration; you may not want either part of the if)
This is it!
Edit: As of AGP 4.2.0, jar.name doesn't return anything useful. For the time being, I used this fragile workaround:
if (jar.file.directoriesInsideRootProject().contains("cats")) {
aspectPath.add(jar.file)
}
fun File.directoriesInsideRootProject() = sequence {
var file = this#directoriesInsideRootProject
while (true) {
yield(file.name)
file = file.parentFile ?: break
if (file == rootProject.projectDir) break
}
}

Scope of variable defined in for loop header

I noticed that the following Kotlin code compiles and executes successfully:
for (i in 1..2) {
val i = "a"
print(i)
}
This prints aa. However, I failed to find rationale behind the decision to allow this kind of variable shadowing. I would say that this is not a good practice, and is prohibited even in Java.
I think that Kotlin designers did a great work of improving Java syntax and accommodating it to the everyday practical use, so I must be missing something here?
Kotlin does not restrict variable shadowing in any way. The rationale is simple: "consistency."
Since you could shadow variables in most other places why would you exclude only some loop variables from the allowed options? Why would they be so special? It is an arbitrary difference.
Any scope can shadow a variable used in another scope. It is NOT good practice and does produce a compiler warning -- but it is allowed.
If you want to engage in a dialog with the contributors of the project, try the discussion forum or slack channel, both are linked from the Kotlin Community page. Otherwise if you feel it is a bug please add an Issue report to Kotlin YouTrack and the answer you receive there will be definitive as well.
In the meantime, you are free to write nonsensical code such as:
val i = 1
class Foo() {
val i = "monkey"
init { println(i) }
#Test fun boo() {
println(i)
val i = i.length
println(i)
if (i == 6) {
val i = Date(System.currentTimeMillis() + i) // Shadow warning
println(i)
}
for (i in 0..i) { // Shadow warning
val i = "chimp $i" // Shadow warning
println(i)
}
InnerFoo()
}
class InnerFoo() {
val i: Long = 100L
init { println(i) }
}
}
Which in Kotlin 1.0.3 produces 3 warnings.
Warning:(15, 21) Kotlin: Name shadowed: i
Warning:(18, 18) Kotlin: Name shadowed: i
Warning:(19, 21) Kotlin: Name shadowed: i
And outputs:
monkeymonkey6Sun Jul 17 11:31:23 UYT 2016chimp 0chimp 1chimp 2chimp 3chimp 4chimp 5chimp 6100