Fully functional mock JSON server with great extras
Mokksy is a mocking tool that lets you build, test and share feature complete Javascript application, even if your API doesn't exist using user-friendly JSON files to mock data.
Mokksy is heavily inspired by the JSON-Server, but Mokksy is build on top of Fastify, when JSON-Server on Express. Mokksy also contains more features like response templates and JWT endpoit protection. I started working on it on the late evenings during the global lockdown in 2020 as a "pet" project,to tinker with the CLI and play with Typescript.
If you like the package, star it on Github, tell your friends about it. If you have any idea for future functionality, feel free to contact me, or raise an issue on Github.
Finally, if you like my code and looking for contract developer to help you deliver high quality JavaScript product - I am available for contract work. I'm UK based.
Installation
The easiest way to enjoy Mokksy is to install it globally via NPM or Yarn. Alternatively, if you want to run it without installing you can run it via NPX.
npm install -g mokksy
yarn global add mokksy
Sometimes, depending on your OS, you might need to install package as root/Administrator. On Linux and MacOS you can do it by prefixing the command with sudo. On Windows you are on your own, I didn't used this OS for years now.
If you don't want to install Mokksy globally, you can install it locally as a Dev dependency: npm install --save-dev mokksy or yarn add -D mokksy. Then please add a script in your package.json section, ie: mokksy: mokksy. This way you will have access to Mokksy via: npm run mokksy or yarn run mokksy. To run Mokksy via NPM directly, without installing it, use NPX command:
npx mokksy db.json
But bear in mind that it will be slower than installed version as package needs to be re-downloaded and re-installed each time.
Getting started, aka. preparing the database file
After installing the server, you need to create a db.json database file (file name is not important, you can call it whatever you like), something like:
{
"posts: [
{"id": 1, "title": "This is my first post"},
{"id": 2, "title": "This is my second post"},
],
"comments": [
{"id": 1, "postId": 1, "body": "This is the very cool first comment"}
],
"profile": {
"name": "My Name", "location": "Earth"
}
}
Alternatively, if your database file is publicly available on the internet (maybe you are sharing it with your work colleagues) you can use the URL instead of file path.
Starting simple server
To start Mokksy, run following command in your terminal:
mokksy db.json
mokksy https://example.com/path/to/your/database.json
First one for local file, second one is to run database available online.
This command will start simple server. Mokksy will try to run on port 5000. If this port will be used by other app, Mokksy will find other free port. You will see all available endpoints printed in the terminal, like in the screenshot at the top of this Readme.
API prefixing
By default API routes are not prefixed, but you can use --api switch to provide the prefix, ie /api to get /api/posts instead of simple /posts. Please remember that prefix needs to start with /.
Customise the server
To check available options, to customise your server, type mokksy --help.
This will print list of available options:
mokksy [options] <sourceFile>
Positionals:
sourceFile JSON database file path [string] [required]
Options:
--help, -h Show this help page [boolean]
--version, -v Show version number [boolean]
--port, -p Set port [default: 5000]
--apiUrlPrefix, --api Prefix the URL path, ie. '/api' for '/api/posts'. Path must start with
'/' [string] [default: ""]
--routes, -r List of custom routes as JSON key:value. On request of the key route
you will be redirected to value. [string] [default: ""]
--idKey, -i Set database 'ID' key (ie. '_id' for Mongo-like collections)
[string] [default: "id"]
--foreignKeySuffix, --fks Set foreign key suffix.
Ie. '_id' for keys like 'user_id', 'post_id', etc.
Default is 'Id' for keys like 'userId', 'postId', etc. [default: "Id"]
--staticPath, -s Set static files directory [string] [default: "public"]
--noStatic, --ns Disable static file server [boolean] [default: false]
--noCors, --nc Disable Cross-Origin Resource Sharing [boolean] [default: false]
--filtering, -f Query params filtering type:
Inclusive (incl) - when element needs to match all filters.
Exclusive (excl) - when the element needs to match just one of the
filters. [string] [choices: "incl", "excl"] [default: "incl"]
--delay, -d Add delay to responses (ms) [number] [default: 0]
--noToken, --nt Disable JWT token endpoint [boolean] [default: false]
--tokenEndpoint, --te URL for your app to process JWT token requests
[string] [default: "/oauth/token"]
--tokenSecret, --ts Secret used to sign tokens on token endpoint. This password is VERY
weak and should be used for testing purpuses only!
[string] [default: "MoKKsy"]
--tokenExpiry, --tex Time in seconds for JWT token to expire. 1 hour by default.
[number] [default: 3600]
--protectEndpoints, --pe Comma separated list of endpoints that needs to be protected by JWT
token.
By default all endpoints are NOT protected. [string] [default: ""]
--template, -t Template file path to format the response data. No template by default.
[string] [default: ""]
Examples:
mokksy --nc -p 8080 db.json Start 'db.json' database on port 8080 and disable CORS.
I know, this is a lot of options and your command can become very, very long. So, as alternative, you can create .mokksyrc or .mokksyrc.json file with configuration. This way you can reduce you command to mokksy db.json. Mokksy will check if config file exists and will load it automatically.
Database snapshot
To save database snapshot press s and then Enter any time, when server is running. This will save database snapshot to your current directory. Files are named mksy-{current-timestamp}.json.
To view current database snapshot, simply go to /_db.
Data filtering, sorting and pagination
You can use query params to filter, search or paginate the data. We are compatible with JSON-server. Those query params can be joined togther to narrow down your search.
Filtering
Simply add query parms coresponding to keys in the collection. You can use . operator to access deep keys, check second example.
GET /posts?title=mokksy
GET /posts?author.name=mac
Filtering operators
You can use special operators _gte (greater than or equal), _lte (lower than or equal), _ne (not equal) or _like by appending it to the end of the search key:
GET /posts?id_gte=5
GET /posts?id_gte=5&id_lte=10
GET /posts?id_ne=5
GET /posts?name_like=jo
Sorting
Use special query params _sort and _order.
GET /posts?_sort=views&_order=asc
Pagination
Use special query param _page and optional _limit. Links to the next pages will be available the the Link response header.
GET /posts?_page=7&_limit=20
Slicing
Use special query param _start, _end or _limit. Total count will be available in the X-Total-Count response header.
Expanding and embeding, aka relationships
To include child resources in the response, use special query param _embed
GET /posts?_embed=comments
To display parent resource in child, add _expand query param in the URL
GET /comments?_expand=post
Display templates
If you don't want to display just a list of plural resources listing, but wrap the resonse, you can use template. You will need to prepare the template file. Response must be valid JSON Object. Here is some example:
{
"object": "list",
"clarkKent": "superman",
"nested": {
"test": {
"dateTime": "{{dateTime}}",
"deeper": {
"why": {
"not": {
"we": {
"can": {
"go": {
"crazy": {
"deep": "{{url}}"
}
}
}
}
}
}
}
}
},
"url": "{{url}}",
"total": "{{total}}",
"timestamp": "{{timestamp}}",
"dateTime": "{{dateTime}}",
"data": "{{data}}"
}
As per example above, you can use some variables, that will be replaced with actual values:
{{dateTime}}will be replaced with current date and time{{url}}will be replaced with currently requested URL{{total}}will be replaced with the number reperesenting total number of items in the response{{timestamp}}will be replaced with Unix timestamp{{data}}will be replaced with actual key listing
Then, run Mokksy with -t switch, for example:
mokksy db.json -t template.json
Token protection and Token endpoint
You can protect the keys from unauthorised access. To do so, please provide the comma-separated list of keys, ie:
mokksy db.json --pe posts,comments
will protect all requests to posts and comments endpoints (get, post, put, delete). You will see 401 error whrn you will try to access http://localhost:5000/posts without the token.
How to get the token?
Token endpoint is available by default at http://localhost:5000/oauth/token, but you can customise it.
As Mokksy is a mock solution, our implementation of the server is very simple. JWT tokens are signed using string ('MoKKsy' by default, customisable using --ts switch).
To get new token, send data payload as POST request to token endpoint (as JSON), and whole request body will become payload of the JWT token.
curl -X "POST" "http://localhost:5000/oauth/token" \
-H 'Content-Type: application/json; charset=utf-8' \
-d $'{
"email": "me@example.com",
"userId": "1"
}'
Token is valid 3600 seconds === 1hour.
How to access the protected resources?
Like in the normal API, add the Authorization header with the Bearer token.
curl "http://localhost:5000/posts" \
-H 'Authorization: Bearer {{your-token-goes-here}}'
Static file server
Mokksy can be also used as a static file server. How? Simply create public folder and start Mokksy from that directory.
Alternatively, use -s switch to set different folder for static files directory.
Static file server is always mapping files to root path /.
CORS
Cross Origin Resource Sharing is enabled by default. It can be disabled using --nc.
Custom routes
You can provide the list of custom routes in the file, in JSON format where key is the final URL and value is the URL where we should redirect your request to.
mokksy db.json -r routes.json
routes.json file should look something like this:
{
"/me": "/profile"
}
At the moment this is super simple, static solution. If you need something more dynamic, let me know what's your use case.
Licence
Mokksy is available under MIT licence.