Smarter snapshot utility for Mocha and BDD test runners + data-driven testing!
This tool makes snapshot testing for Mocha (and other BDD)
frameworks quick and painless. This module spies on global it
function,
which allows it to accurately get test information (beating static code parsing
done in snap-shot); it should work in transpiled code.
This package uses snap-shot-compare to display object and text difference intelligently.
This function also includes data-driven testing mode, similar to sazerac, see Data-driven testing section below.
Requires Node version 4 or above.
npm install --save-dev snap-shot-it
Example from spec.js
const snapshot = require('snap-shot-it')
describe('example', () => {
it('works', () => {
snapshot(add(10, 20))
snapshot('a text message')
return Promise.resolve(42).then(snapshot)
})
})
Run Mocha tests, then open the created snapshots/spec.js file
exports['example works 1'] = 30
exports['example works 2'] = "a text message"
exports['example works 3'] = 42
Suppose you change the resolved value from 42
to 80
const snapshot = require('snap-shot-it')
describe('example', () => {
it('works', () => {
snapshot(add(10, 20))
snapshot('a text message')
return Promise.resolve(80).then(snapshot)
})
})
The test will fail
1) example works:
Error: 42 !== 80
The error message should intelligently handle numbers, objects, arrays, multi-line text, etc.
The returned value includes saved value (after any transformations) and saved snapshot name. Usually it is spec name + index, or could be exact name
const out = snapshot('my name', 42)
// {value: 42, key: 'my name'}
You can see the saves snapshot values by running with environment variable
SNAPSHOT_SHOW=1 npm test
You can see snapshot values without writing them into the snapshot file
SNAPSHOT_DRY=1 npm test
You can update snapshot values
SNAPSHOT_UPDATE=1 npm test
You can use the following aliases: SNAPSHOT_UPDATE=1
, SNAPSHOTS_UPDATE=1
or SNAP_SHOT_UPDATE=1
.
If you want to sort saved snapshots alphabetically inside each snapshot file, run with
SNAPSHOT_SORT=1 npm test
You can also set the config option in the package.json file
{
"config": {
"snap-shot-it": {
"sortSnapshots": true
}
}
}
Hopefully sorting snapshots would help when updating them.
Renaming tests might lead to confusion and pruning snapshots. You can name the snapshots yourself
const value = 42
snapshot('my name', value)
The snapshots will be saved as
exports['my name'] = 42
Note you should make sure that the name is unique per spec file.
If you do want to share a named snapshot value from several places or tests in the same spec file, you need to pass an option when calling snapshot
. The the first snapshot is saved, and the next ones will just compare against the value.
snapshot('my shared snapshot', value, { allowSharedSnapshot : true })
// some time later
snapshot('my shared snapshot', value, { allowSharedSnapshot : true })
If the test run is successful and executed all tests (there was no .only
) then snapshots without a test are pruned. You can skip pruning by running with environment variable
SNAPSHOT_SKIP_PRUNING=1 npm test
Writing multiple input / output pairs for a function under test quickly becomes tedious. Luckily, you can test a function by providing multiple inputs and a single snapshot of function's behavior will be saved.
// checks if n is prime
const isPrime = n => ...
it('tests prime', () => {
snapshot(isPrime, 1, 2, 3, 4, 5, 6, 7, 8, 9)
})
The saved snapshot file will have clear mapping between given input and produced result
// snapshot file
exports['tests prime 1'] = {
"name": "isPrime",
"behavior": [
{
"given": 1,
"expect": false
},
{
"given": 2,
"expect": true
},
{
"given": 3,
"expect": true
},
{
"given": 4,
"expect": false
},
{
"given": 5,
"expect": true
},
...
]
}
You can also test functions that expect multiple arguments by providing arrays of inputs.
const add = (a, b) => a + b
it('checks behavior of binary function add', () => {
snapshot(add, [1, 2], [2, 2], [-5, 5], [10, 11])
})
Again, the snapshot file gives clear picture of the add
behavior
// snapshot file
exports['checks behavior of binary function add 1'] = {
"name": "add",
"behavior": [
{
"given": [
1,
2
],
"expect": 3
},
{
"given": [
2,
2
],
"expect": 4
},
{
"given": [
-5,
5
],
"expect": 0
},
{
"given": [
10,
11
],
"expect": 21
}
]
}
See src/data-driven-spec.js for more examples.
Run with environment variable DEBUG=snap-shot-it ...
to see log messages.
Because under the hood it uses snap-shot-core you might
want to show messages from both libraries with DEBUG=snap-shot* ...
You can pass your own NPM modules as pre-compare
, compare
and store
functions using package.json
. For example, to use both local and 3rd party NPM modules
{
"config": {
"snap-shot-it": {
"pre-compare": "./pre-compare",
"compare": "snap-shot-compare",
"store": "./store"
}
}
}
Each NPM module in this case should export a definition of a function that matches the expected core function
pre-compare
is simply an identity or transformation functioncompare
should match snap-shot-core#compare-function, for example see snap-shot-comparestore
is another identity or transformation function, see snap-shot-core#store-function
By default, all snapshots are stored in the same folder __snapshots__
, which can lead to name clashes. You can set an option in your package.json file to create a nested folder structure inside __snapshots__
folder that mimics the spec structure. Use config > snap-shot-it
object in the package.json file.
{
"config": {
"snap-shot-it": {
"useRelativePath": true
}
}
}
input spec files
specs/
spec.js
subfolder/
spec2.js
result should be
__snapshots__/
specs/
spec.js
subfolder/
spec2.js
An example using ts-mocha is shown in folder ts-demo
CoffeeScript example is in coffee-demo folder. Watch mode is working properly.
Came during WorkBar Cambridge Happy Hour on the terrace as I was thinking about
difficulty of adding CoffeeScript / TypeScript support to
snap-shot project. Got the idea of overriding global.it
when
loading snap-shot
because a day before I wrote repeat-it
which overrides it and it is very simple repeat/src/index.js.
This NPM module is part of my experiments with snapshot testing. There are lots of other ones, blog posts and slides on this topic.
- snap-shot-core implements loading and saving snapshots
- snap-shot is an alternative to this package that tries to determine spec name using stack trace and static source code parsing. Hard to do for transpiled code!
- schema-shot is "schema by example" snapshot testing
- subset-shot where new value can be a superset of the saved snapshot
- Blog post Picking snapshot library
- Slides Snapshot testing the hard way
Author: Gleb Bahmutov <[email protected]> © 2017
License: MIT - do anything with the code, but don't blame me if it does not work.
Support: if you find any problems with this module, email / tweet / open issue on Github
Copyright (c) 2017 Gleb Bahmutov <[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.