It is a relational json db inspired by typeorm. It is an alternate to full-fledged relational db which typeorm supports. Purpose of this db is to be able to store a json or encrypted json file as a db remotely, avoiding the side-effects of ephemeral nature of server environment.
This DB is recommended to use for your nestjs/nodejs: lightweight/hobby projects, prototypes, portfolios etc.
Below are the steps to use this library.
This step is about configuring the DB. There are two mandatory methods downloadDbJson()
and downloadDbJson()
you need to provide to ready the db system.
Below is an example:
import { JsonDB } from 'json-relational-db';
import { downloadDbJson, uploadDbJson } from './sync-file';
JsonDB.configure({
downloadDbJson: downloadDbJson,
uploadDbJson: uploadDbJson,
});
Below is an example of downloadDbJson()
and downloadDbJson()
methods/functions. It uses google-cloud storage, however, you can use any storage or local file:
import { Storage } from '@google-cloud/storage';
const storage = new Storage();
const fileName = 'db.json';
const bucketName = 'my-bucket';
export async function downloadDbJson(): Promise<object> {
const res = await storage.bucket(bucketName).file(fileName).download();
const object = JSON.parse(res.toString());
return object;
}
export async function uploadDbJson(object: object): Promise<void> {
await storage
.bucket(bucketName)
.file(fileName)
.save(JSON.stringify(object));
}
This step is about defining entities/tables. Below is an example of creating an entity/table
import { Entity, Column, ColumnType } from 'json-relational-db';
import { projectEntity } from './project.entity';
export interface Skill {
skill_id: number;
name: string;
description: string;
proficiency: string;
experience_in_month: number;
proficiency_level: string;
}
export const skillEntity = new Entity<Skill>('skill', [
new Column('skill_id', ColumnType.SERIAL),
new Column('name', ColumnType.STRING),
new Column('description', ColumnType.STRING),
new Column('proficiency', ColumnType.NUMBER),
new Column('experience_in_month', ColumnType.NUMBER),
new Column('proficiency_level', ColumnType.STRING),
new Column('project_id', projectEntity), //foreign key
]);
This step is about using the entity repository. Below is a sample service class which makes use of the repository methods:
import { Skill, skillEntity } from 'src/entities';
export class SkillService {
private readonly skillRepo = skillEntity.repository;
async getAllSkills(): Promise<SkillDTOWithId[]> {
const skills = await this.skillRepo.find({
orderBy: [{ column: 'proficiency', asc: false }],
});
return skills.map((s) => this.mapSkillFromEntity(s));
}
async addSkill(skills:Skill[]) {;
return await this.skillRepo.save(skills);
}
async updateSkill(skill: Skill) {
const s = await this.skillRepo.findOne({
skill_id: skill.skillId,
});
if (!s) {
throw new Error('Skill not found');
}
await this.skillRepo.update(
{
skill_id: s.skill_id,
},
skill
);
return skill;
}
}
Below example shows how to make use of relations. In this example we are using 3 tables and querying the related data:
async getMyProfile(
query: GetMyProfile.Query
): Promise<MyProfileFullDTOWithId> {
const profileId = +query.profileId;
const myProfile = await this.myProfileRepo.findOne({
my_profile_id: profileId,
});
if (myProfile == null) {
throw new HttpException('Profile not found', HttpStatus.BAD_REQUEST);
}
const location = await this.locationRepo.find({
where: { my_profile_id: profileId },
});
const professionalProfile = await this.professionalProfileRepo.find({
where: {
my_profile_id: profileId,
},
});
return this.mapProfileDTOFromEntity(
myProfile,
location,
professionalProfile
);
}
Github: https://github.com/emreSlim/json-db
Feel free to contact me or raise an issue about this package.