gulp-karma2

gulp-karma

Karma plugin for gulp 3

Usage

First, install gulp-karma as a development dependency:

npm install --save-dev gulp-karma

Then, add it to your gulpfile.js:

var karma = require('gulp-karma');
 
var testFiles = [
  'client/todo.js',
  'client/todo.util.js',
  'client/todo.App.js',
  'test/client/*.js'
];
 
gulp.task('test', function() {
  // Be sure to return the stream
  return gulp.src(testFiles)
    .pipe(karma({
      configFile: 'karma.conf.js',
      action: 'run'
    }))
    .on('error', function(err) {
      // Make sure failed tests cause gulp to exit non-zero
      throw err;
    });
});
 
gulp.task('default', function() {
  gulp.src(testFiles)
    .pipe(karma({
      configFile: 'karma.conf.js',
      action: 'watch'
    }));
});

API

karma(options)

options.configFile

Type: String

The path to the Karma configuration file.

options.action

Type: String
Default: run

One of the following:

• run: Start the server, run tests once, then exit.
• watch: Start the server, run tests once, then watch for changes and run when files change.

options.*

Any Karma option can be passed as part of the options object. See Karma Configuration for a complete list of options. Note: It's best practice to put options in your Karma config file.

Notes

Task return value

Karma runs asynchronously. When using action: 'run' in a task, you should return the stream so gulp knows the task finished.

Watching

Due to the way Karma works, using gulp.watch to watch files results in contrived usage that doesn't work as expected in some cases. As a result, Karma's watch mechanism is employed to make usage of this plugin as straight forward as possible.

Globs

Globs are resolved before they're sent to Karma, so if you add a new file that matches a glob you passed using gulp.src('test/*').pipe(karma), it won't be caught by Karma.