Indra, continuous integration and automation tool

Indra allows you to define continuous integration and automation scripts as Dart code. Scripts can be ran from an IDE and debugged. Indra runs as a command line tool or as a daemon with a JSON API. Indra has a user interface that interacts with the daemon.

Installation

First, install Dart.

Then, clone Indra from GitHub:

git clone https://github.com/stijnvanbael/indra.git ~/.indra/runtime

Finally, install Indra:

Linux / MacOS:

cd ~/.indra/runtime/bin
./install.sh

Windows:

• Create an environment variable INDRA_HOME = <path\to\indra>\bin
• Add INDRA_HOME to your PATH environment variable

Now you can run any Indra script as

indra <script.dart> [param1=value [param2=value [...]]]

Or in case the file is named build.dart, simply

indra <[param1=value [param2=value [...]]]

Example

Example script: reflective.dart

import 'dart:isolate';

import 'package:indra/src/cli.dart';
import 'package:indra/src/tasks/git.dart';
import 'package:indra/src/tasks/pub.dart';

main(List<String> args, SendPort outputPort) async {
  var params = setup(outputPort, args, defaultParams: {'branch': 'dev'});

  var git = new GitRepo('git@github.com:stijnvanbael/reflection.git', branch: params['branch']);
  await git.cloneOrPull();
  await Pub.get();
  await Pub.run('test/reflective.dart');
}

To run the script

indra reflective.dart

Daemon

Indra can also be ran as a daemon:

indra-daemon <working/dir>

Any Dart file under the working directory can be scheduled to run from the daemon by calling:

POST http://localhost:8080/jobs/<name of the file excluding .dart>/schedule

You can request which jobs are running by calling:

GET http://localhost:8080/jobs

You can request the output of a job by calling:

GET http://localhost:8080/jobs/<script name>/<sequence number>/output