Good Vibes is a simple Node.js testing library designed to make writing and executing tests easy and fun.
It was inspired by, and written primarily while listening to, Alicia Keys' Tiny Desk Concert
npm install good-vibes --save-dev
Lets say you wanted to test the following function defined in code.js
// defined in ./src/code.js
const codeToTest = (firstName, lastName, age) => {
return {
name: {
first: firstName,
last: lastName,
},
age: age
}
}
export codeToTest;
// use the following if you use commonjs style exports
// exports.codeToTest = codeToTest
Here is how you would write a test using good-vibes for this function
// defined in ./test/code.test.js
import { codeToTest } from "./src/code";
import run, { test } from "good-vibes";
// alternatively `const { test, run } = require('good-vibes');
test("My first test", (context) => {
const expected = {
name: { first: "Hello", last: "World" },
age: 999,
};
context.log("Lets do this!");
context.check(expected, codeToTest("Hello", "World", 999));
context.done();
});
run(); // runs all your tests defined using the `test` api
To run this test add a script to your package.json
file
{
"name": "my-library",
"version": "0.0.1",
"scripts": {
"test": "node test/code.test.js"
}
}
And then on command line type
npm test
The test
function allows you to define a test. It has the following signature:
test("name", testFunction, "groupName");
Name of your test
Expects a function with the following signature:
const testFunction = (context) => {
// run your code
// make assertions using context.check(expected, actual)
// mark test as complete using context.done()
};
A simple utility and assertion framework wrapped within the context of your test and group that provides the following api's:
check(expected, actual)
: uses lodash.isEqual to perform deep equality checks on primitives and objects and moredone()
: marks test as completelog(message)
: log provides a simple wrapper overconsole.log
with the test name prefixed to your message to make them easier to find in the logssnapshot(name, actual, updateBaseline)
: allows for snapshot testing, discussed later in this guide
Allows you to create a group of tests. More on this below. If not specified the test is assiged to the Default
group.
Calling the run()
api starts the test execution. It accepts a configuration object as follows:
run({
timeout: 300_000, // in milliseconds
snapshotsDirectory: "./test/__snapshots/", // string
executionTimePrecision: 4, // number
});
This is the total amount of time all tests have to run before good-vibes end execution.
Default is set to 5 minutes
. After 5 minutes good-vibes will end all test execution and return with error code 1
Default value is set to ./test/__snapshots__/
. For more details see section on Snapshot Testing
below.
This option specifies the precision to which execution time of the tests are calculated . It is an optional value which is set to a default value of 3
.
TODO
All tests defined using test
are considered to be asynchronous function. This is the reason you need to tell good-vibes that your test is complete by calling the context.done()
api.
Here is an example of an asynchronous test
// defined in ./async.test.js
import run, { test } from "good-vibes";
const sayGreeting = async (message) => {
return new Promise((resolve, reject) => {
setTimeout(() => resolve(`Hello, ${message}`), 2000); // resolve value after 2 seconds
});
};
test("Async Test", async (context) => {
// note the async declaration
const actual = await sayGreeting("World!"); // waiting for result
context.check("Hello, World!", actual);
context.done(); // mark test as complete
});
run(); // runs all your tests defined using the `test` api
The test
api supports groupName
as the third argument which allows you to groups similar tests together. Groups have the following features:
- Groups always run sequentially one after the other
- All tests inside the group must finish before next group starts
- Tests inside a group run concurrently (unless
sync
is called)
- Groups can have setup and teardown code defined using
before
andafter
api
import run, { before, test, after } from "good-vibes";
const MY_GROUP = "My Group";
let numbers;
let strings;
before((context) => {
// context inside before and after blocks only have done() and log(message) api's
numbers = [1, 2, 3, 4, 5];
context.log("Logging is supported inside before and after blocks as well");
setTimeout(() => {
strings = ["this", "value", "is", "set", "after", "2", "seconds"];
context.done(); // dont forget to call done
}, 2000);
}, MY_GROUP);
test(
"My Test",
(ctx) => {
ctx
.check(5, numbers.length)
.check("1,2,3,4,5", numbers.join(","))
.check(7, strings.length)
.check("this value is set after 2 seconds", strings.join(" "))
.done();
},
MY_GROUP
);
after((ctx) => {
numbers = undefined;
strings = undefined;
ctx.done();
}, MY_GROUP); // group name is important so dont forget it :)
run();
before
function runs once before all tests and has the following signature
before(beforeFunction, groupName); // groupName defaults to 'Default' group
beforeFunction
has the following signature
const beforeFunction = async (context) => {
// perform your test setup here
// call context.done() once complete
};
after
function runs once after all tests have finished and has the same signature as before
It can sometimes be verbose to have to specify the group name in each of before
, after
and test
api calls. To make this easier good-vibes provides the group
api which wraps all api's with the specified group name.
Lets rewrite the above example now using the group
api.
import run, { group } from "good-vibes";
const MY_GROUP = "My Group";
const { before, test, after } = group(MY_GROUP); // wrap good-vibes api's with MY_GROUP groupName
let numbers;
let strings;
before((ctx) => {
numbers = [1, 2, 3, 4, 5];
setTimeout(() => {
strings = ["this", "value", "is", "set", "after", "2", "seconds"];
ctx.done();
}, 2000);
});
test("My Test", (ctx) => {
ctx
.check(5, numbers.length)
.check("1,2,3,4,5", numbers.join(","))
.check(7, strings.length)
.check("this value is set after 2 seconds", strings.join(" "))
.done();
});
after((ctx) => {
numbers = undefined;
strings = undefined;
ctx.done();
});
run();
Sometimes the output of a function can be very large - making it cumbersome to test using the context.check
api. To help with this good-vibes supports snapshot testing.
Snapshot tests allow you test for changes in the expected output. They do this by maintaining a baseline
file with the expected output. Then whenever you run your test the actual output is compared against this baseline file and if they don't match the differences are reported as shown below:
import { group } from "good-vibes";
const { test } = group("Snapshots");
let createLargeObject = (value) => ({
a: value,
b: 123,
c: true,
d: [1, 2, 3, 4, 5],
e: {
f: [
{
a: value,
b: 123,
c: true,
d: [1, 2, 3, 4, 5],
},
],
},
});
test("Large Object", (ctx) => {
// snapshot verifications are asynchronous so you must await its execution
await ctx.snapshot("Check 1", createLargeObject("One"));
// A single test may have one or more snapshot assertions, each snapshot inside a test must have a unique name
await ctx.snapshot("Check 2", createLargeObject("Two"));
// dont forget to call done() at the end
ctx.done();
});
run();
In the above example good-vibes
will check if the output of createLargeObject
matches the snapshot with name Check 1
and Check 2
respectively.
Notes:
- Snapshots internally use JSON.stringify to write the snapshot files, so please be careful while creating snapshots of objects with functions in them
- Order of keys in the object does not matter as the verification is performed on the parsed JSON object and not the json string itself
The snapshot
function allows you to define a test. It has the following signature:
async snapshot<Type>(snapshotName: string, actualValue: Type, rebase?: boolean): Promise<TestContext>
The first time you run the snapshot test it will fail stating that the baseline file could not be found, similar to the image below:
To create a new baseline or update an existing one, set the third argument of the snapshot
api to true
Important Note: When you rebase a snapshot file, good-vibes will automatically fail that test. This is to safeguard against accidental updates to snapshot files and prevent this flag from being committed into source control
By default good-vibes will create a directory to store all snapshots at this location: ./test/__snapshots__
. Inside this folder a new folder is created for each group
of tests.
You may override this property by setting the snapshotsDirectory
configuration property while calling the run
function. See section on run
api above for more details.
All snapshot files are named using the following convention: <test-name>_<snapshot-name>.json
Putting the above conventions together, the following template specifies the exact location (using defaults): ./test/__snapshots__/<group-name>/<test-name>_<snapshot-name>.json
In case of our example tests above, following would be the snapshot paths:
./test/__snapshots__/Snapshots/Large Object_Check 1.json
./test/__snapshots__/Snapshots/Large Object_Check 2.json
By default all tests inside a group are considered async
and run concurrently. However you may change this behavior at the group level by use of the sync
api. The order of the tests in synchronous testing mode is the order of declaration in the test file from top to bottom.
sync(groupName?: string) // groupName defaults to "Default"
import run, { before, test, sync } from "good-vibes";
const MY_GROUP = "My Group";
let numbers;
let strings;
before((context) => {
numbers = [1, 2, 3, 4, 5];
strings = ["sample", "values"];
}, MY_GROUP);
test(
"My Test",
(ctx) => {
ctx.check(5, numbers.length).check("1,2,3,4,5", numbers.join(",")).done();
},
MY_GROUP
);
// You can invoke the sync api anywhere in any of your test files
// Because tests only run when the `run()` api is called
// good-vibes will ensure all tests in this group run synchonously
sync(MY_GROUP);
test(
"My Test",
(ctx) => {
ctx
.check(7, strings.length)
.check("sample values", strings.join(" "))
.done();
},
MY_GROUP
);
run();
import run, { group } from "good-vibes"; // import sync from good-vibes
const MY_GROUP = "My Group";
const { before, test, sync } = group(MY_GROUP);
sync();
let numbers;
let strings;
before((context) => {
numbers = [1, 2, 3, 4, 5];
strings = ["sample", "values"];
});
test("My Test", (ctx) => {
ctx.check(5, numbers.length).check("1,2,3,4,5", numbers.join(",")).done();
});
test("My Test", (ctx) => {
ctx.check(7, strings.length).check("sample values", strings.join(" ")).done();
});
run();
When you have too many tests in your application, its often useful to be able to run just one or a few failing tests alone to debug the cause of their failure.
good-vibes support this using a special purpose group name called Debug
. Debug
group name can be used in before
, test
and after
api's and even inside concise groups.
Recommendation: We recommend that you import { DEBUG } from 'good-vibes'
instead of redefining the variable yourself. This will isolate your tests from changes to this group name in the future.
Important Note: When running in debug mode good-vibes exits with failure return code 1 - this is to prevent the debug flag from being accidentally committed to source control. The following disclaimer will also be printed in the logs:
const run, { before, test, DEBUG } = require("../lib/index");
// When running in debug mode only tests tagged with DEBUG group are run,
// all other tests are skipped
// Good vibes also exits with a return code of 1 to prevent tests from being
// checked into your codebase accidentally
before((ctx) => {
ctx.log("You can run before and after blocks even in debug mode");
ctx.done();
}, DEBUG);
test(
"Johnny Cash",
(ctx) => {
setTimeout(() => {
ctx.log("Hurt");
ctx.check(1, 1).done();
}, 2000);
},
DEBUG
);
run();
import run, { group, DEBUG } from "good-vibes";
const MY_GROUP = "My Group";
const { before, test, sync } = group(MY_GROUP);
let numbers;
let strings;
before((context) => {
numbers = [1, 2, 3, 4, 5];
strings = ["sample", "values"];
}, DEBUG);
// always ensure that you run your before blocks in DEBUG mode too
// if you before some setup for the test being debugged
test(
"My Test",
(ctx) => {
ctx.check(5, numbers.length).check("1,2,3,4,5", numbers.join(",")).done();
},
DEBUG
); // running only one test in DEBUG mode
test("My Test", (ctx) => {
ctx.check(7, strings.length).check("sample values", strings.join(" ")).done();
});
run();
Code coverage reports can be generated using the nyc
package. For up-to-date instruction please check out Installation & Usage section of nyc
First install nyc
as a dev dependency
npm install nyc --save-dev
Next update your package.json
scripts section to add a coverage script to run nyc
{
"name": "my-library",
"version": "0.0.1",
"scripts": {
"test": "node test/code.test.js",
"coverage": "nyc --reporter=lcov --reporter=text-summary npm run test"
}
}
The above configuration generates a text-summary and a detailed lcov report with html visualization
This can be found inside the /coverage
folder, a sample report is shown below:
- Simple tests can be found in test/simple.test.js
- Test grouping with use of
before
andafter
can be found in test/groups.test.js - Example of
snapshot
tests can be found in test/snapshot.test.js - A concise example of groups using the
group
api can be found in test/concise-groups.test.js - Examples of
sync
tests can be found in test/synchronous.test.js - Examples of failing tests can be found in test/failing.test.js
- Example of how
debug
mode works can be found in test/debug.test.js