Home • Explore • Blog • Models • CI
☰



nixpkgs
2nd chunk of `pkgs/applications/editors/jetbrains/readme.md`
503dfabbc9700d69c781704c1525f7f1820ddb15dfa35eb30000000100000cb5
 - The list can also contain drvs giving the directory contents of the plugin (this is how you use a plugin not added to nixpkgs) or a single `.jar` (executable). For an example, look at the implementation of `fetchPluginSrc` in `plugins/default.nix`.

### How to add a new plugin to nixpkgs
 - Find the page for the plugin on https://plugins.jetbrains.com
 - Find the id (it's the number after https://plugins.jetbrains.com/plugin/)
 - Run `plugins/update_plugins.py` add (plugin id)
 - If binaries need patch or some other special treatment, add an entry to `plugins/specialPlugins.nix`

## How to update stuff:
 - Run ./bin/update_bin.py, this will update binary IDEs and plugins, and automatically commit them
 - Source builds need a bit more effort, as they **aren't automated at the moment**:
   - Run ./source/update.py ./source/ides.json ./bin/versions.json. This will update the source version to the version of their corresponding binary packages.
   - Run these commands respectively:
     - `nix build .#jetbrains.idea-community-src.src.src && ./source/build_maven.py source/idea_maven_artefacts.json result/` for IDEA
     - `nix build .#jetbrains.pycharm-community-src.src.src && ./source/build_maven.py source/pycharm_maven_artefacts.json result/` for PyCharm
   - Update `brokenPlugins` timestamp and hash (from https://web.archive.org/web/*/https://plugins.jetbrains.com/files/brokenPlugins.json)
   - Do a test build
     - Notice that sometimes a newer Kotlin version is required to build from source, if build fails, first check the recommended Kotlin version in `.idea/kotlinc.xml` in the IDEA source root
     - Feel free to update the Kotlin version to a compatible one
   - If it succeeds, make a commit
   - Run ./plugins/update_plugins.py, this will update plugins and automatically commit them
   - make a PR/merge
   - If it fails, ping/message GenericNerdyUsername or the nixpkgs Jetbrains maintainer team

## How to add an IDE:
 - Make dummy entries in `bin/versions.json` (make sure to set the version to something older than the real one)
 - Run `bin/update_bin.py`
 - Add an entry in `bin/ides.json`
 - Add an entry in `default.nix`

### TODO:
 - move/copy plugin docs to nixpkgs manual
 - replace `libxcrypt-legacy` with `libxcrypt` when supported
 - make `jetbrains-remote-dev.patch` cleaner
 - is extraLdPath needed for IDEA?
 - set meta.sourceProvenance for everything
 - from source builds:
   - remove timestamps in output `.jar` of `jps-bootstrap`
   - automated update scripts
   - fetch `.jar` s from stuff built in nixpkgs when available
     - what stuff built in nixpkgs provides `.jar`s we care about?
       - kotlin
   - make `configurePhase` respect `$NIX_BUILD_CORES`
   - make the subdir of the resulting tar.gz always have a release number (2023.2.2) instead of a build number (232.9921.89)
 - jdk:
   - build on darwin
   - use chromium stuff built by nixpkgs for jcef?
   - make `buildPhase` respect `$NIX_BUILD_CORES`
   - automated update script?
 - on `aarch64-linux`:
   - test plugins
   - from source build
   - see if build (binary or source) works without expat
 - on `x86_64-darwin`:
   - test plugins
   - from source build
 - on `aarch64-darwin`:
   - test plugins
   - from source build
Title: JetBrains Plugin, IDE Management, and Development TODOs
Summary
This document outlines procedures for managing JetBrains plugins and IDEs within the Nixpkgs environment. It details adding new plugins via ID and script, and updating components. Binary IDEs and plugins can be updated automatically using `bin/update_bin.py`. Source builds are manual, involving source version updates, Maven artifact builds for IDEs (e.g., IDEA, PyCharm), `brokenPlugins` data updates, test builds (possibly adjusting Kotlin), and a final `plugins/update_plugins.py` run. Adding new IDEs requires updating multiple configuration files. A comprehensive TODO list covers moving plugin docs, library replacements, automating source builds, improving JAR fetching, JDK build enhancements, and expanding testing across architectures (aarch64-linux, x86_64-darwin, aarch64-darwin).