Difference between revisions of "Eclipse plugin package guidelines"

From ArchWiki
Jump to navigation Jump to search
(→‎Sample PKGBUILD: fixed stupid error that prevented features from being installed if already unpacked)
Line 30: Line 30:
   # Features
   # Features
   find features -type f -name *.jar | while read _feature ; do
   find features -type f | while read _feature ; do
     if [[ ${_feature} =~ (.*\.jar$) ]] ; then
     if [[ ${_feature} =~ (.*\.jar$) ]] ; then
       install -dm755 ${_dest}/${_feature%*.jar}
       install -dm755 ${_dest}/${_feature%*.jar}

Revision as of 16:47, 28 June 2009


There are many ways to install working Eclipse plugins, especially since the introduction of the dropins directory in Eclipse 3.4, but some or them are messy, and having a standardized and consistent way of packaging is very important to lead to a clean system structure. It's not easy, however, to achieve this without the packager knowing every detail about how Eclipse plugins work. This page aims to define a standard and simple structure for Eclipse plugin PKGBUILDs, so that the filesystem structure can remain consistent between all plugins without having the packager to start again for every new package.

Eclipse plugin structure and installation

The typical Eclipse plugin contains two directories, features and plugins, and since Eclipse 3.3 they could only be placed in /usr/share/eclipse/. The content of these two directories could be mixed with that of other plugins, but it created a mess and rendered the structure difficult to manage. It was also very difficult to tell at a glance which package contained which file.

This installation method is still supported in Eclipse 3.4, but the preferred one is now using the /usr/share/eclipse/dropins/ directory. Inside this directory can live an unlimited number of subdirectories, each one containing its own features and plugins subdirectories. This allows to keep a tidy and clean structure, and should be the standard packaging way.



Here's an example, we'll detail how to customize it below.

pkgdesc="A task-focused interface for Eclipse"
arch=('i686' 'x86_64')
optdepends=('bugzilla: ticketing support')

build() {

  cd ${srcdir}

  # Features
  find features -type f | while read _feature ; do
    if [[ ${_feature} =~ (.*\.jar$) ]] ; then
      install -dm755 ${_dest}/${_feature%*.jar}
      cd ${_dest}/${_feature/.jar}
      jar xf ${srcdir}/${_feature} || return 1
      install -Dm644 ${_feature} ${_dest}/${_feature}

  # Plugins
  find plugins -type f | while read _plugin ; do
    install -Dm644 ${_plugin} ${_dest}/${_plugin}

How to customize the build

The main variable which needs to be customized is the pkgname. If you are packaging a typical plugin, then this is the only thing you need to do: most plugins are distributed in zip files which only contain the two features and plugins subdirectories. So, if you're packaging the foo plugin and the source file only contains the features and plugins, you just need to change pkgname to eclipse-foo and you're set.

Read on to get to the internals of the PKGBUILD, which help to understand how to setup the build for all the other cases.

In-depth PKGBUILD review

Package naming

Packages should be named eclipse-pluginname, so that they're recognizable as eclipse-related packages and it's easy to extract the plugin name with a simple shell substitution like ${pkgname/eclipse-}, not having to resort to an unneded ${_realname} variable. The plugin name is necessary to tidy up everything during installation and to avoid conflicts.



Some plugins need the features to be extracted from jar files. The jar utility, already included in the JRE, is used to do this. However, jar can't extract to directories other than the current one: this means that, after every directory creation, it's necessary to cd inside it before extracting. The ${_dest} variable is used in this context to improve readability and PKGBUILD tidiness.


As we said, source archives provide two directories, features and plugins, each one packed up with jar files. The preferred dropins structure should look like this:


This structure allows for mixing different versions of libraries that may be needed by different plugins while being clear about which package owns what. It will also avoid conflicts in case different packages provide the same library. The only alternative would be splitting every package from its libraries, with all the extra fuss it requires, and it wouldn’t even be guaranteed to work because of packages needing older library versions. Features have to be unjarred since Eclipse won't detect them otherwise, and the whole plugin installation won’t work. This happens because Eclipse treats update sites and local installations differently (don't ask why, it just does).

The build() function

First thing to be noticed is the cd ${srcdir} command. Usually source archives extract the features and plugins folders directly under ${srcdir}, but this is not always the case. Anyway, for most non-(de facto)-standard plugins this is the only line that needs to be changed. Next is the features section. It creates the necessary directories, one for every jar file, and extracts the jar in the corresponding directory. Similarly, the plugins section installs the jar files in their directory. A while cycle is used to prevent funny-named files.