acnsy
v2.0.1
Published
rethinked async
Downloads
6
Maintainers
Readme
rAsync - async rethinked
This module is a wrapper to the popular async module, providing an easy way to combine serial/waterfall and parallel task execution. With this you can define a workflow of function calls using nested arrays. It helps to your code keep simple and readable.
Usage
var rAsync = require('r-async');
rAsync(initialData /* object (optional) */, tasks /* array */, finally /* function */);
The r-async
module gives you a single function, which requires three parameters:
- initialData : object (optional) - you can pass initial data to your tasks, this will be passed to the first codes to be executed as parameter (see below)
- tasks : array - pass the functions, that you would like to execute. The given funcions will execute in series, while nesting further arrays of function calls will alternate between parallel and series execution, as you go more deep in nesting (see the example)
- finally : function - pass a function here, which gets executed after the tasks, whether there was any errors or everything run fine
The tasks
Each task must be defined in the following form, recieving two parameters
function(params /* object */, callback /* function */){ ... }
- params : object - parameters recieved from previously executed tasks or initialData
- callback : function - you must execute this, to mark, that the task is finished. If you don't call the callback the process is breaked and the other tasks doesn't calls.
The callback must be called with two parameters:
callback(error /* null or any */, params /* object */)
- error : Any (for example Error object) or null - if there was any errors during the execution of the task, then this should recieve the Error object, or any truthy value which describing the error, else it should be null
- params : object - the recieved, and optionally altered params object should be sent back to rAsync through this
Defining serial and parallel execution
Let's look at the task definitions through an example. Supposingly we have five functions, as seen below:
function A(params, callback){
console.log('A');
callback(null, params);
}
function B(params, callback){
console.log('B');
callback(null, params);
}
function C(params, callback){
console.log('C');
callback(null, params);
}
function D(params, callback){
console.log('D');
callback(null, params);
}
function E(params, callback){
console.log('E');
callback(null, params);
}
By default, if we just dump these into an array, then rAsync will execute these in series:
var tasks = [A, B, C, D, E]; // A -> B -> C -> D -> E
rAsync({}, tasks, function(){});
If we introduce an array in the array of tasks, then it's contents will execute in parallel:
var tasks = [A, [B, C], D, E]; // A --> B ---> D -> E
\-> C -/
Another level of nesting will provide us with changing back to serial execution and further nesting will keep shifting from the two modes.
var tasks = [A, [B, [C, D], E]; // A ---> B ----> E
\-> C -> D -/
It is not possible to change the initial mode of execution, but we can always wrap the tasks in multiple arrays and there is no limit for how deep the nesting can go.
var tasks = [[A, B, C, D, E]]; // all executed in parallel
Conditional execution
//...require your modules
var rAsync = require('r-async');
rAsync({}, [
yourFirstTask,
rAsync.ifTask(yourConditionTask, yourTrueBranchTask, yourElseBranchTask)
],function(err,params){
console.log(err,params)
});
Where yourConditionTask
must fill the ifresult
property in the params. For example:
function yourConditionTask(params, callback){
params.ifresult = params.xy==42; //calculate your result of condition
callback(null, params);
}