Rivet is a task-based build tool, scripted in JavaScript and executed using Node.js. Similar to make or rake, it can be used in any scenario that can be broken into tasks, from building native and web applications to automating deployment procedures.
$ npm install -g rivet
From the command line, simply use rivet to execute a task.
$ rivet hello
Tasks are declared in a rivet file, by default named "tasks.js".
Tasks are declared using rivet.task(), and given a name, optional
prerequisites, and a function to execute.
module.exports = function(rivet) {
rivet.desc('say hello')
rivet.task('hello', function() {
console.log('Hello!');
});
}
All rivet files export a function using standard module.exports boilerplate.
For brevity, this is omitted from further examples.
Asynchronous tasks are as easy as accepting a done callback and invoking it
when the task completes.
rivet.task('hello', function(done) {
setTimeout(function() {
console.log('Hello! (in a little while)');
done();
}, 1000);
});
In many cases, a task often requires that another task execute first. These are known as prerequisites. Rivet ensures that all prerequisites have been executed prior to any task that requires them.
rivet.task('hello', 'lookup_name', function() {
console.log('Hello ' + this.scratch.name + '!');
});
rivet.task('lookup_name', function() {
this.scratch.name = 'Dave';
});
Also demonstrated here is what's known as the "scratch". This is a shared area that any task can write to or read from, making it convenient to pass data between tasks.
Tasks can be grouped into namespaces, making it easy to organize related tasks and limit there scope in larger projects.
rivet.namespace('formal', function() {
rivet.task('hello', function() {
console.log('Hello, sir!');
});
rivet.task('greet', 'hello', function() {
console.log('How may I be of assistance?');
});
});
Prerequisites resolve within the current namespace. Relative namespaces can be used to refer to parent namespaces if necessary.
rivet.namespace('informal', function() {
rivet.task('greet', '^:hello', function() {
console.log('What can I help you with?');
});
});
The use of a caret (^) is used to refer to a parent namespace. These can be
strung together as needed. For example, ^:^:hello would reference the hello
task in the grandparent namespace.
A single task can be declared multiple times. In this case, each function is additive. When the task is executed, each step will be invoked in sequence.
rivet.task('archive', function(done) {
copy(['app.js', 'utils.js'], 'output', done);
})
rivet.task('archive', function(done) {
zip('output', 'output.zip', done);
})
Rivet provides syntactic sugar in the form of "targets" and "steps", which lets this be expressed in a form that is more clear.
rivet.target('archive', function() {
this.step(function(done) {
copy(['app.js', 'utils.js'], 'output', done);
})
this.step(function(done) {
zip('output', 'output.zip', done);
})
});
This is an effective way to break up a set of asynchronous operations, writing them as if they were sequential commands.
Rivet is conceptually the same as Jake. I was using Jake for this purpose, but ultimately found it lacking for the following reasons:
- Jake treats asynchronous tasks as second-class, requiring extra options to enable them. This works against the Node.js grain.
- Tasks are not "additive", in that they can't be redeclared. I found that this tended to result in unnecessary verbosity and nested callbacks.
- Prerequisites don't resolve within the containing namespace. This limited the effectiveness of using namespaces and resulted in parent namespaces being peppered throughout separate files.
All of these things could (and should, IMO) be fixed in Jake. However, after attempting to do that, I concluded that it would be simpler to start from a fresh codebase where other simplifying assumptions could be made.
Grunt does away with prerequisites and centralizes configuration while tying that configuration to the type of task. While this approach certainly works, it doesn't suit my own personal tastes.
I prefer to decouple the implementation of tasks from their configuration. By embracing the functional aspects of JavaScript, and setup functions popularized by Connect and Express middleware, it is easy to write succinct tasks with declarative, inline configuration. For example:
function zip(dir, zipfile) {
return function(done) {
var command = 'zip -r ' + zipfile + ' ' + dir;
exec(command, done)
}
}
task('zip_app', zip('app', 'app.zip'));
task('zip_plugins, zip('plugins', 'plugins.zip'));
While completely subjective, I find this much more aesthetically pleasing. This syntax also makes it easy to retain prerequisites, which are too useful to be unsupported by a build tool.
$ npm install --dev
$ make test
(The MIT License)
Copyright (c) 2012 Jared Hanson
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.