Alexander Udalov bba25042b6 Fix generation of metadata for const val field in multi-file class
Metadata for declarations in multi-file classes is always written to the
corresponding parts, but fields for public const vals are generated into
the facade class.

Before this change, we ran generateBackingField only to generate the
const val field in the facade (because isBackingFieldOwner is true only
for the facade). Besides generating the actual field,
generateBackingField also stores the mapping to the FIELD_FOR_PROPERTY
slice, which is used later in the serializer to serialize the correct
JVM signature of the field to the metadata. However, the mapping was
stored to the incorrect binding map! Namely, the map from the same
ClassBuilder that is used to generate the class, and that is the facade
class builder. But the metadata needs to be generated to the part class,
not the facade.

Surprisingly, there is a test at
reflection/multifileClasses/javaFieldForVarAndConstVal.kt that checks
that kotlin-reflect is able to load the field signature of a const val
in a multi-file class, which is the main use case of this information,
and it passed before this change. The reason the test passed is that
KPropertyImpl.javaField uses JvmProtoBufUtil.getJvmFieldSignature which
mistakenly treats the absence of the field signature in the proto as the
hint that it may be computed naively (just the property's name + naively
mapped type), whereas in fact the field signature was missing
completely. (The fact that absence of a message may mean two things is
certainly a shortcoming of the metadata format and should be revisited
in the future.) This never led to any problem in the real world because
it's easy to prove that the field signature of a const val can always be
computed naively.

After this change, we now run generateBackingField twice: for the
facade's class builder, and for the part's class builder. The actual
field is only generated in the former case. But the mapping computed in
the latter case is used correctly in JvmSerializerExtension to store the
field signature to metadata. Also refactor
generateSyntheticMethodIfNeeded in the similar way, to reduce the number
of `if`s.
2018-08-16 14:11:09 +02:00
2018-08-07 15:59:43 +03:00
2018-05-21 18:19:49 +03:00
2018-08-03 21:29:19 +03:00
2018-03-02 03:15:28 +03:00
2018-08-02 12:14:58 +02:00
2017-11-01 17:09:10 +03:00
2016-11-15 16:42:57 +01:00
2018-08-07 12:21:56 +03:00
2017-06-12 19:20:12 +02:00

official project TeamCity (simple build status) Maven Central GitHub license

Kotlin Programming Language

Welcome to Kotlin! Some handy links:

Editing Kotlin

Build environment requirements

In order to build Kotlin distribution you need to have:

  • JDK 1.6, 1.7, 1.8 and 9

  • Setup environment variables as following:

      JAVA_HOME="path to JDK 1.8"
      JDK_16="path to JDK 1.6"
      JDK_17="path to JDK 1.7"
      JDK_18="path to JDK 1.8"
      JDK_9="path to JDK 9"
    

For local development, if you're not working on bytecode generation or the standard library, it's OK to have only JDK 1.8 and JDK 9 installed, and to point JDK_16 and JDK_17 environment variables to your JDK 1.8 installation.

You also can use Gradle properties to setup JDK_* variables.

Note: The JDK 6 for MacOS is not available on Oracle's site. You can download it here.

Building

The project is built with Gradle. Run Gradle to build the project and to run the tests using the following command on Unix/macOS:

./gradlew <tasks-and-options>

or the following command on Windows:

gradlew <tasks-and-options>

On the first project configuration gradle will download and setup the dependencies on

  • intellij-core is a part of command line compiler and contains only necessary APIs.
  • idea-full is a full blown IntelliJ IDEA Community Edition to be used in the plugin module.

These dependencies are quite large, so depending on the quality of your internet connection you might face timeouts getting them. In this case you can increase timeout by specifying the following command line parameters on the first run:

./gradlew -Dhttp.socketTimeout=60000 -Dhttp.connectionTimeout=60000

Important gradle tasks

  • clean - clean build results
  • dist - assembles the compiler distribution into dist/kotlinc/ folder
  • ideaPlugin - assembles the Kotlin IDEA plugin distribution into dist/artifacts/Kotlin folder
  • install - build and install all public artifacts into local maven repository
  • runIde - build IDEA plugin and run IDEA with it
  • coreLibsTest - build and run stdlib, reflect and kotlin-test tests
  • gradlePluginTest - build and run gradle plugin tests
  • compilerTest - build and run all compiler tests
  • ideaPluginTest - build and run all IDEA plugin tests

OPTIONAL: Some artifacts, mainly Maven plugin ones, are built separately with Maven. Refer to libraries/ReadMe.md for details.

Building for different versions of IntelliJ IDEA and Android Studio

Kotlin plugin is indented to work with several recent versions of IntelliJ IDEA and Android Studio. Each platform is allowed to have a different set of features and might provide a slightly different API. Instead of using several parallel Git branches, project stores everything in a single branch, but files may have counterparts with version extensions (*.as32, *.172, *.181). The primary file is expected to be replaced with its counterpart when targeting non-default platform.

More detailed description of this scheme can be found at https://github.com/JetBrains/bunches/blob/master/ReadMe.md.

Usually, there's no need to care about multiple platforms as all features are enabled everywhere by default. Additional counterparts should be created if there's an expected difference in behavior or an incompatible API usage is required and there's no reasonable workaround to save source compatibility. Kotlin plugin contains a pre-commit check that shows a warning if a file has been updated without its counterparts.

Development for some particular platform is possible after 'switching' that can be done with Bunch Tool from the command line.

cd kotlin-project-dir

# switching to IntelliJ Idea 2018.2
bunch switch . 182

Working with the project in IntelliJ IDEA

Working with the Kotlin project requires IntelliJ IDEA 2017.3. You can download IntelliJ IDEA 2017.3 here.

To import the project in Intellij choose project directory in Open project dialog. Then, after project opened, Select File -> New... -> Module from Existing Sources in the menu, and select build.gradle.kts file in the project's root folder.

In the import dialog, select use default gradle wrapper.

To be able to run tests from IntelliJ easily, check Delegate IDE build/run actions to Gradle and choose Gradle Test Runner in the Gradle runner settings.

At this time, you can use the latest released 1.2.x version of the Kotlin plugin for working with the code. To make sure you have the latest version installed, use Tools | Kotlin | Configure Kotlin Plugin Updates and press "Check for updates now".

Compiling and running

From this root project there are Run/Debug Configurations for running IDEA or the Compiler Tests for example; so if you want to try out the latest and greatest IDEA plugin

  • VCS -> Git -> Pull
  • Run the "IDEA" run configuration in the project
  • a child IntelliJ IDEA with the Kotlin plugin will then startup

Including into composite build

To include kotlin compiler into composite build you need to define dependencySubstitution for kotlin-compiler module in settings.gradle

includeBuild('/path/to/kotlin') {
    dependencySubstitution {
        substitute module('org.jetbrains.kotlin:kotlin-compiler') with project(':include:kotlin-compiler')
    }
}

Contributing

We love contributions! There's lots to do on Kotlin and on the standard library so why not chat with us about what you're interested in doing? Please join the #kontributors channel in our Slack chat and let us know about your plans.

If you want to find some issues to start off with, try this query which should find all Kotlin issues that marked as "up-for-grabs".

Currently only committers can assign issues to themselves so just add a comment if you're starting work on it.

A nice gentle way to contribute would be to review the standard library docs and find classes or functions which are not documented very well and submit a patch.

In particular it'd be great if all functions included a nice example of how to use it such as for the hashMapOf() function. This is implemented using the @sample macro to include code from a test function. The benefits of this approach are twofold; First, the API's documentation is improved via beneficial examples that help new users and second, the code coverage is increased.

Some of the code in the standard library is created by generating code from templates. See the README in the stdlib section for how to run the code generator. The existing templates can be used as examples for creating new ones.

Also the JavaScript back-end could really use your help. See the JavaScript contribution section for more details.

If you want to contribute a new language feature, it is important to discuss it through a KEEP first and get an approval from the language designers. This way you'll make sure your work will be in line with the overall language evolution plan and no other design decisions or considerations will block its acceptance.

Submitting patches

The best way to submit a patch is to fork the project on GitHub and then send us a pull request via GitHub.

If you create your own fork, it might help to enable rebase by default when you pull by executing

git config --global pull.rebase true

This will avoid your local repo having too many merge commits which will help keep your pull request simple and easy to apply.

Checklist

Before submitting the pull request, make sure that you can say "YES" to each point in this short checklist:

  • You provided the link to the related issue(s) from YouTrack
  • You made a reasonable amount of changes related only to the provided issues
  • You can explain changes made in the pull request
  • You ran the build locally and verified new functionality
  • You ran related tests locally and they passed
  • You do not have merge commits in the pull request
S
Description
The Kotlin Programming Language.
Readme 2.1 GiB
Languages
Kotlin 79.9%
Java 10.4%
Swift 4.3%
C 2.8%
C++ 2.1%
Other 0.3%