should is an expressive, readable, test framework agnostic, assertion library for node.
It extends the Object prototype with a single non-enumerable getter that allows you to express how that object should behave.
should literally extends node's assert module, in fact, it is node's assert module, for example should.equal(str, 'foo') will work, just as assert.equal(str, 'foo') would, and should.AssertionError is asset.AssertionError, meaning any test framework supporting this constructor will function properly with should.
var user = {
name: 'tj'
, pets: ['tobi', 'loki', 'jane', 'bandit']
};
user.should.have.property('name', 'tj');
user.should.have.property('pets').with.lengthOf(4)
someAsyncTask(foo, function (err, result) {
should.not.exist(err);
should.exist(result);
result.bar.should.equal(foo);
});
$ npm install should
As mentioned above, should extends node's assert. The returned object from require('should') is thus similar to the returned object from require('assert'), but it has one extra convenience method:
should.exist('hello')
should.exist([])
should.exist(null) // will throw
This is equivalent to should.ok, which is equivalent to assert.ok, but reads a bit better. It gets better, though:
should.not.exist(false)
should.not.exist('')
should.not.exist({}) // will throw
We may add more assert extras in the future... ;)
should's assertion chaining provides an expressive way to build up an assertion, along with dummy getters such as an, have, and be, provided are what I am simply calling modifiers, which have a meaning effect on the assertion. An example of this is the not getter, which negates the meaning, aka user.should.not.have.property('name'). In the previous example note the use of have, as we could omit it and still construct a valid assertion.
Some modifiers such as include only have an effect with specific assertion methods, for example when asserting a substring like so: str.should.include.string('test'), we could omit include, but it helps express the meaning, however keys has a strict effect, unless the include modifier is used.
Some assertions can be chained, for example if a property is volatile we can first assert property existence:
user.should.have.property('pets').with.lengthOf(4)
which is essentially equivalent to below, however the property may not exist:
user.pets.should.have.lengthOf(4)
our dummy getters such as and also help express chaining:
user.should.be.a('object').and.have.property('name', 'tj')
The returned object from require('should') is the same object as require('assert'). So you can use should just like assert:
should.fail('expected an error!')
should.strictEqual(foo, bar)
In general, using the Object prototype's should is nicer than using these assert equivalents, because should gives you access to the expressive and readable language described above:
foo.should.equal(bar) // same as should.strictEqual(foo, bar) above
The only exception, though, is when you can't be sure that a particular object exists. In that case, attempting to access the should property may throw a TypeError:
foo.should.equal(bar) // throws if foo is null or undefined!
For this case, require('should') extends require('assert') with an extra convenience method to check whether an object exists:
should.exist({})
should.exist([])
should.exist('')
should.exist(0)
should.exist(null) // will throw
should.exist(undefined) // will throw
You can also check the negation:
should.not.exist(undefined)
should.not.exist(null)
should.not.exist('') // will throw
should.not.exist({}) // will throw
Once you know an object exists, you can safely use the should property on it.
Assert truthfulness:
true.should.be.ok
'yay'.should.be.ok
(1).should.be.ok
or negated:
false.should.not.be.ok
''.should.not.be.ok
(0).should.not.be.ok
Assert === true:
true.should.be.true
'1'.should.not.be.true
Assert === false:
false.should.be.false
(0).should.not.be.false
Assert Arguments:
var args = (function(){ return arguments; })(1,2,3);
args.should.be.arguments;
[].should.not.be.arguments;
Asserts that length is 0:
[].should.be.empty
''.should.be.empty
({ length: 0 }).should.be.empty
equality:
({ foo: 'bar' }).should.eql({ foo: 'bar' })
[1,2,3].should.eql([1,2,3])
strict equality:
should.strictEqual(undefined, value)
should.strictEqual(false, value)
(4).should.equal(4)
'test'.should.equal('test')
[1,2,3].should.not.equal([1,2,3])
Assert inclusive numeric range:
user.age.should.be.within(5, 50)
Assert typeof:
user.should.be.a('object')
'test'.should.be.a('string')
Assert instanceof:
user.should.be.an.instanceof(User)
[].should.be.an.instanceof(Array)
Assert numeric value above the given value:
user.age.should.be.above(5)
user.age.should.not.be.above(100)
Assert numeric value below the given value:
user.age.should.be.below(100)
user.age.should.not.be.below(5)
Assert regexp match:
username.should.match(/^\w+$/)
Assert length property exists and has a value of the given number:
user.pets.should.have.length(5)
user.pets.should.have.a.lengthOf(5)
Aliases: lengthOf
Substring assertion:
'foobar'.should.include.string('foo')
'foobar'.should.include.string('bar')
'foobar'.should.not.include.string('baz')
Assert inclusion of object:
var obj = {foo: 'bar', baz: {baaz: 42}};
obj.should.include.object({foo: 'bar'});
obj.should.include.object({baz: {baaz: 42}});
obj.should.not.include.object({foo: 'baz'});
Assert property exists and has optional value:
user.should.have.property('name')
user.should.have.property('age', 15)
user.should.not.have.property('rawr')
user.should.not.have.property('age', 0)
Assert own property (on the immediate object):
({ foo: 'bar' }).should.have.ownProperty('foo')
Assert array value:
[1,2,3].should.contain(3)
[1,2,3].should.contain(2)
[1,2,3].should.not.contain(4)
Assert own object keys, which must match exactly, and will fail if you omit a key or two:
var obj = { foo: 'bar', baz: 'raz' };
obj.should.have.keys('foo', 'bar');
obj.should.have.keys(['foo', 'bar']);
using the include modifier, we can check inclusion of a key, but not fail when we omit a few:
obj.should.include.keys('foo')
obj.should.include.keys('bar')
obj.should.not.include.keys('baz')
Assert that the given property is a function:
user.should.respondTo('email')
For example you can use should with the Expresso TDD Framework by simply including it:
var lib = require('mylib')
, should = require('should');
module.exports = {
'test .version': function(){
lib.version.should.match(/^\d+\.\d+\.\d+$/);
}
};
To run the tests for should simple update your git submodules and run:
$ make test
Yes, yes it does, with a single getter should, and no it wont break your code, because it does this properly with a non-enumerable property.
(The MIT License)
Copyright (c) 2010-2011 TJ Holowaychuk <[email protected]> Copyright (c) 2011 Aseem Kishore <[email protected]>
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.