DAL

Data Access Layer DAL is a proxy layer for SQL databases with a MongoDB inspired query interface. It can be used as a Go or NodeJS package (requires compiler) to create your own proxy and apply custom middlewares.

Notes:

• This project is still in early alpha. You need to build it yourself and use at your own risk.
• At the time only SQLite is implemented, however, other drivers might work.

Use cases:

• For IOT networks when MySQL/PG are too heavy.
• If you need a layer between your application and the database (i.e. for caching).
• If you want a MongoDB-like query interface for your SQL db.
• When you need a SQLite proxy (useful to share datasets with services)

Usage

Server

The most efficient way to use DAL is to run the server as a standalone service.

Build:

go build -o server

Run:

export SQLITE_DIRECTORY=/opt/data

./server

2024/08/21 22:01:54 Starting server on port 8118
2024/08/21 22:01:54 Using directory: /opt/data

Client

Install:

pnpm add git+git@github.com:nesterow/dal.git

Query Interface

Method	Description	SQL
In(table: string)	Select table	SELECT * FROM table
Find(filter: object)	Filter rows	SELECT * FROM table WHERE filter
Fields(fields: string[])	Select fields	SELECT fields, FROM table
Sort(sort)	Sort rows	SELECT * FROM table ORDER BY sort
Limit(limit: number)	Limit rows	SELECT * FROM table LIMIT limit
Offset(offset: number)	Offset rows	SELECT * FROM table OFFSET offset
Join({ $for: "t_2", $do: { "t.a": "b" } })	Join tables	SELECT * FROM table t JOIN t_2 ON t.a = b
Insert({name: "J"}, {name: "B"})	Insert row	INSERT INTO table (name,) VALUES ('J', 'B')
Set({name: "Julian"})	Update row (Find(filter).Set({}))	UPDATE table SET name = 'Julian' WHERE filter
Delete()	Delete row (Find(filter).Delete())	DELETE FROM table WHERE filter
As(DTO)	Map rows to a DTO	SELECT * FROM table
Rows()	Get rows iterator	SELECT * FROM table
Exec()	Execute query (update, insert, delete)	SQL RESULT
Query()	Query database	DTO array
Tx()	Run in trasaction

Filters

Filter	Description	SQL
{id: 1, num: 2}	Equals, default filter	WHERE id = 1 AND num = 2
{id: { $eq: 1 }}	Equals, explicit	WHERE id = 1
{id: { $gt: 1 }}	Greater than	WHERE id > 1
{id: { $gte: 1 }}	Greater than or equal	WHERE id >= 1
{id: { $lt: 1 }}	Less than	WHERE id < 1
{id: { $lte: 1 }}	Less than or equal	WHERE id <= 1
{id: { $ne: 1 }}	Not equal	WHERE id != 1
{id: { $in: [1, 2] }}	In	WHERE id IN (1, 2)
{id: { $nin: [1, 2] }}	Not in	WHERE id NOT IN (1, 2)
{id: { $like: "a" }}	Like	WHERE id LIKE '%a%'
{id: { $nlike: "a" }}	Not like	WHERE id NOT LIKE '%a%'
{id: { $between: [1, 2] }}	Between	WHERE id BETWEEN 1 AND 2
{id: { $nbetween: [1, 2] }}	Not between	WHERE id NOT BETWEEN 1 AND 2
{id: { $glob: "\*son" }}	Glob	WHERE id GLOB '*son'

Example

import { DAL } from "@nesterow/dal";

class UserDTO {
  id: number = 0;
  name: string = "";
  data: string = "";
  age: number | undefined;
}

const db = new DAL({
  database: "test.sqlite",
  url: "http://localhost:8111",
});

// SELECT * FROM test t WHERE name GLOB '*son' AND age >= 18
const rows = db
  .In("test t")
  .Find({
    name: { $glob: "*son" },
    age: { $gte: 18 },
  })
  .As(UserDTO) // Map every row to DTO
  .Rows();

for await (const row of rows) {
  console.log(row); // Jason, Jackson
}

Internals

The client uses a light builder and messagepack over http. It is relatively easy to implement a client in any language see the docs

License

While in alpha stage the project is free for research purposes. Later it will be released under MIT-like license with AI/dataset exclusion terms.