π§ Under development π§
This project is under heavy development and not yet suitable for production use!
Deta Base ORM is a JavaScript/TypeScript library which lets your work with Deta Base in a object relational model. Define a schema for your Base, cross reference items between Bases and interact with your data in a more functional way. It is heavily inspired by mongoose and helps you write high quality, maintainable applications in a more productive way then with the standard Deta Base SDK (it uses it under the hood). Additionally it features a offline mode which mocks the remote Deta Base and replaces it with a local JSON file for better access during development.
First be sure you have Node and NPM installed.
Next install Deta Base ORM from the command line:
npm install deta-base-orm --save
Then import it into your project:
import * as DetaOrm from 'deta-base-orm'
Or:
const DetaOrm = require('deta-base-orm')
Here's a quick example to help you get started!
Imagine we like kittens and want to record every kitten we ever meet in Deta Base:
import * as DetaOrm from 'deta-base-orm'
// β¨ Define a schema for the kittens
const KittenSchema = new DetaOrm.Schema({
name: 'string',
cuteness: 'number'
})
// π Create our Kitten base
const Kitten = new DetaOrm.Base('Kitten', KittenSchema)
// π± Create a new kitten
const line = Kitten.create({
name: 'Line',
cuteness: 8
})
// π Access the kittens name
console.log(line.name) // 'Line'
// πΎ Save our kitten to the Deta Base
await line.save()
// π Find all kittens
const kittens = await Kitten.find()
console.log(kittens) // [{name: 'Line', cuteness: 8}, ...]
// π Find a kitten by its key
const sameKitten = await Kitten.findByKey(line.key)
console.log(sameKitten) // {name: 'Line', cuteness: 8}
// 𧡠Find a kitten by its cuteness
const cutest = await Kitten.find({ cuteness: 8 })
console.log(cutest) // [{name: 'Line', cuteness: 8}]
// π Delete a kitten
await sameKitten.delete()
Pretty cool right? Congratulations, you now know how almost everything works! π
See below for a more detailed guide.
Before you can properly use the other methods you need to define a schema for your Base. It specifies what values to expect and in what format. You can set a property to required (false
by default) or specify a default value:
const schema = new DetaOrm.Schema({
name: 'string',
age: {
type: 'number',
required: true
},
hungry: {
type: 'boolean',
default: false
}
})
Creating a new Base is similar to how you would do it with the regular Deta Base SDK except that it also accepts a schema. The schema can either be a instance of the Schema class or an object defining a schema (like above):
const Base = new DetaOrm.Base('name', schema, options)
Creating a new document/item in your Base is very easy. Just call the .create()
method of your Base instance and pass it some new data. Make sure the data matches your schema, else it will throw and error telling you what's wrong.
const document = Base.create({
name: 'Maxi',
age: 5,
hungry: true
})
Creating a document doesn't save it to Deta Base automatically. It only exists locally until you call
.save()
To save a document to your Deta Base, call .save()
:
await document.save()
You can also create and save a document in one go by calling .save()
on your Base instance instead of .create()
:
const document = await Base.save({
name: 'Maxi',
age: 5,
hungry: true
})
There are multiple ways to retrieve documents from your Base:
const documents = await Base.find(query)
The query should be a JavaScript object specifing attributes to filter for defined in the schema. See Filtering below.
const document = await Base.findOne(query)
The query should be a JavaScript object specifing attributes to filter for defined in the schema. See Filtering below.
await document.update(changes)
The changes should be a JavaScript object defining the changes.
await document.delete()
You can cross reference documents from one Base to another:
// πββοΈ Declare Base for humans
const Human = new DetaOrm.Base('Human', {
name: 'string'
})
// π Declare Base for kittens
const Kitten = new DetaOrm.Base('Kitten', {
name: 'string',
owner: Human
})
// πΆ Create new human
const human = await Human.save({
name: 'Maxi'
})
// π± Create new kitten
const cat = await Kitten.save({
name: 'Line',
owner: human.key // Reference id of human
})
console.log(cat) // => { name: 'Line', owner: '17be07f6ee7zNXWw' }
// β¨ Replace id of the owner with actual data from another Base
await cat.populate('owner')
console.log(cat) // => { name: 'Line', owner: { name: 'Maxi' } }
.populate()
takes a path to a field in your schema and returns a document matching the value of that path from another Base. It only stores the key of the document in the Base, the acutal data is resolved from the other Base. For this to work you need to define the other Base in your schema:
const Human = new DetaOrm.Base('Human', {
name: 'string'
})
const Kitten = new DetaOrm.Base('Kitten', {
name: 'string',
owner: Human, // This tells the library that this path is a cross reference to another Base
// Does the same as above
owner: {
type: 'base',
base: Human
}
})
When getting a document from a base with .find()
and .findOne()
you can specify a query object or list of queries to filter the documents with. In the object you can select paths from your schema and define filters and queries to match it against:
const Kitten = new DetaOrm.Base('Kitten', {
name: 'string',
cuteness: 'number'
})
const garfield = await Kitten.findOne({
name: 'Garfield' // Name equals value
})
const cutest = await Kitten.find({
name: {
$con: 'e' // All kittens with the letter e in their name
},
cuteness: {
$gt: 8 // All kittens with a cuteness greater than 8
}
})
This library supports all of Deta Bases queries:
Name | Property | Description |
---|---|---|
Equal | $eq | Equal to value |
Not Equal | $ne | Not equal to value |
Less Than | $lt | Less than number |
Greater Than | $gt | Greater than number |
Less Than Or Equal | $lte | Less than or equal to number |
Greater Than Or Equal | $gte | Greater than or equal to number |
Prefix | $pfx | Prefix of value |
Range | $rg | Range of numbers i.e. [22,30] |
Contains | $con | String contains substring or array contains item |
Not Contains | $ncon | Opposite of contains |
They can be combined together in one query or used in different queries. All queries in an object are interpreted as AND
:
await Base.find({
name: {
$con: 'repo',
$pfx: 'gh'
}
})
Here: name starts with the prefix
gh
AND contains the substringrepo
Queries in separate objects are treated as OR
:
await Base.find([
{
name: 'Hello'
},
{
cuteness: {
$gt: 8
}
}
])
Here: The name must be
Hello
OR the cuteness greater than8
Deta Base ORM supports the same paging as the normal Base SDK. Just specify a limit and optionally the key of the last item as parameters: .find(query, limit, last)
Example:
const items = await Base.find({ age: 24 }, 5) // Limit number of items to 5
const nextFive = await Base.find({ age: 24 }, 5, last: items[4].key) // Use the key of the last item
During the execution of your Deta Micro there are a few different environment variables available which give your information about the micro itself and the environment it is running in. For ease of use, deta-base-orm
provides them and a few other useful environment variables to you via the Runtime
class:
import { Runtime } from 'deta-base-orm'
console.log(Runtime.isMicro)
console.log(Runtime.region)
console.log(Runtime.memory)
console.log(Runtime.domain)
Here is an overview of the available variables:
Key | Description | Type | Micro Only |
---|---|---|---|
isMicro | Is the program running in a Deta micro | boolean | No |
isDev | Is the program running in development mode | boolean | No |
isOffline | Is the offline mode enabled | boolean | No |
isSpace | Is the micro running inside Space | boolean | No |
projectKey | Deta project key | string | No |
subdomain | Subdomain of the micro | string | Yes |
domain | Full domain of the micro | string | No |
region | AWS region the micro is running in | string | Yes |
memory | Amount of memory available to the micro | number | Yes |
logLevel | Deta micro log level | boolean | No |
You can optionally enable the offline mode which uses locally stored JSON files instead of the live remote Deta Base service. This is very helpful during development as it allows you to run your queries against a test database and then use the live one in production. And of course, it works offline so you can develop your application with all of the benefits of Base without internet access.
You can enable the offline mode via the environment variable DETA_OFFLINE
or using the Base option offline
:
const OfflineBase = new DetaOrm.Base('name', schema, {
offline: true
})
It will store the JSON files in a
.deta-base-orm
folder by default
You can optionally pass a options object to the the Base contructor:
const options = { /* optional */ }
const Base = new DetaOrm.Base(name, schema, options)
Here are all the available options:
By default this library will generate random keys which are in an ascending order, meaning that new documents will always be at the bottom. You can change this by setting descending
to true. New documents will then be at the top of your Base.
In ascending mode it will use the Unix timestamp and in descending order maxDateNowValue (8.64e15) - Unix timestamp
to make sure the key is sequential. The key will be appended with a random id to make sure two documents created at the same time do not have the same key. The final key will consist of the sequential timestamp in hex plus a 5 char random id.
Because the timestamp is in ms, the key is only sequential until a certain point i.e two keys generated in the same ms may not be in the right order
Set offline
to true to enable the offline mode. In offline mode it will use a locally stored JSON files instead of the live remote Deta Base service.
The JSON files will by default be stored in a .deta-base-orm
directory but this can be changed with the storagePath
option. For each Base you create, a JSON file with the same name will be created in that folder. E.g the data for a Base with the name Kittens will be stored in Kittens.json
.
Important Note: The offline mode is currently still in development and doesn't yet support everything that is available with the normal Base service. Here are all the things that are missing/don't fully work yet:
- Advanced query operators: The query currently only supports direct matching of values i.e.
{ name: 'Maxi', age: 18 }
- Limits/paging: Currently all items are returned that match the query
- Cross referencing: Running
.populate()
to cross reference items across Bases doesn't work and returns an error
This library can optionally add a createdAt
field to each new document containing the timestamp of when it was created. To enabled this, set timestamp
to true.
If you already have a Deta Base instance you can pass it to Deta Base ORM as the db
option to reuse it:
const deta = Deta()
const db = deta.Base(name)
const Base = new DetaOrm.Base(name, schema, { db })
- Add custom methods/actions to the Base
- Make populate work offline
- Add populate as chained command to
find()
- Populate multiple paths at once
- run
yarn lint
ornpm run lint
to run eslint. - run
yarn watch
ornpm run watch
to watch for changes. - run
yarn build
ornpm run build
to produce a compiled version in thelib
folder.
This project was developed by me (@betahuhn) in my free time. If you want to support me:
Copyright 2021 Maximilian Schiller
This project is licensed under the MIT License - see the LICENSE file for details.