This section comprises of the core heart of the ResearchKernel i.e our backend.
We are using three Databases, serving different functionality for our product. Following is the explanation of why those database is necessary for us.
| Database | Database Type | Purpose |
|---|---|---|
| ElasticSearch | NoSQL | We store all the metadata of research papers into the ElasticSearch and we also use this for our Search Engine. This is our core database where all the information of research papers (Title, Abstract, authors,...etc) are stored and served to the website. |
| Neo4j | Graph | We are building the Knowledge Graph of research papers we have in ElasticSearch, this database also stores our recommendation system result. Our short term goal is to combine our Knowledge graph and ElasticSearch that will make our search research more relevant |
| MySQL | SQL | We are planing to provide personalised recommendations and notifications, we are storing all the user data into our MySQL database as of now. See Sequelize Docs |
As we are using node.js for our backend development so proficiency in node.js is a must. We are using three database, we do have diverse tasks and features to implement, you should know at least one database before-hand.
We are more focused on building ElasticSearch and MySQL REST APIs as those databases contains the soul of our product. Neo4j is our second priority as we are not actively working on our graph database.
We share database credentials to our committers only. Below is a small chunk of our all production database dumps and collaborators can use those to build APIs or suggested features.
MySQL, ElasticSearch, Neo4j
To be added.
We have three main branches, and temporary feature branches with short lifespan.
| Branch | Purpose |
|---|---|
| Master | This is the branch we will use in production servers |
| Staging | This is a backup branch for Master and for testing production code. |
| Development | This is the branch where all the feature branches will merge. |
| Features | These branches are for developing new feature that will be integrated to our product, and have a short life span. |
- Select task on Trello
- Inform to your committer and they will assign you the task on Trello
- Create a feature branch from development.
- Push a PR to development.
Need help @Tushar Mudgal
This is a straightforward boilerplate for building REST APIs with ES6 and Express.
- ES6 support via babel
- CORS support via cors
- Body Parsing via body-parser
# clone it
git clone https://github.com/ResearchKernel/backend.git
cd backend
# Install dependencies
npm install
# Start development live-reload server
npm run dev
# Start production server:
npm start- Ubuntu
- Mac
- Windows
- Start MySQL server on your local machine.
- For Mac:
mysql.server start - For Other OS: see appropriate commands.
- For Mac:
- Now create development database using following commands.
- Open MySQL from CLI (Run command
mysql). CREATE DATABASE researchkernel_development.
- Open MySQL from CLI (Run command
- Now open the project in any editor (I prefer VSCode ;)
- Create a new file
.envin project's root. - Copy contents of
.env.exampleinto.env. - Edit database config in
.envaccording to your MySQL config. - Start the node server again.
- You're good to go.
- Create a new file
MIT
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent