the simple stub/spy javascript library
bond only provides stubbing and spy functionality. For test running and assertions, you will need to use other libraries.
Keep up to date with changes by checking the releases.
bond(object, 'propertyOrMethodName') returns the bond api
bond() returns an anonymous spy
bond().return(value) returns an anonymous spy that returns the given value when called
bond#to(value) replaces the value with a new value; reverts the stub after the current test completes
bond#return(value) replaces the value with a spy that returns the given value; reverts the spy after the current test completes
bond#asyncReturn(values...) replaces the value with a spy that calls the last argument passed to the function with the provided values
bond#through() replaces the value with a spy, but allows it to return its normal value
bond#restore() replaces a spy/stub with its original value; useful for implementing your own cleanup handler (see below)
spy.called is a call count for the spy
spy.calledWith(arg1, arg2, ...) is a test for being called with specific values
spy.calledArgs is an array of methods calls, each index holds the array of arguments for that call
npm install bondjs -> bond = require 'bondjs'
<script src="bond.js"> -> window.bond(...)
with mocha, qunit, jasmine: These frameworks should work with bond as is. Bond looks for a global function named either afterEach or testDone to implement its spy/stub restore functionality. If those exist, as they should when using these frameworks, it should work fine.
with some other test runner: You may need to implement your own cleanup method for bond to work properly. This might look like the following.
bond.cleanup = someTestRunner.registerAfterCallback
see the test.coffee file for examples
use npm test to run the tests