Goquiz provides the scrapers to get the engineering multiple-choice questions across the sites, structure the data, populate the database, and Serve API endpoints to retrieve the questions. You can also start the services with db-less mode.
Currently, Goquiz can scrap the MCQ questions from the following sites.
Scrapers for more web sites will be provided as the project progress. There are over 4000+ MCQ questions from 74 different categories and all of them are credited to the respective original web source. This project is educational purpose only.
The project structure was inspired by hexagonal architecture. Actors such as transport and repository are separated from service, which allows service to be a technology-agnostic component with only business logic inside.
The actors are loaded during runtime based on the config, i.e. , postgres data mode and in-memory data model can be swapped easily just by config value. It allows mocking and significantly improves testability.
The scraper spawns go routine to scrap each web page, consolidate and categorized the data before populating to the database.
Goquiz can be served as either REST or gRPC API ( still in development ). PostgresSQL is used as persistent storage to store the categorized data. For easier deployment, db-less mode with in-memory data model can also be used.
- Get the mcq questions from the web sites
- Group the questions into separate categories
- Write the questions for each categories to the Postgres database
- Get categories by category ID endpoint to retrieve all available categories
- Get questions by categories endpoint to retrieve all questions from one category
- Key based API authentication
- API rate limiting
Download and install go on your machine and clone the goquiz project. For deployment, you can deploy without database or setup and populate the database first before serving the API. With later option, you don't have to scrap the web everytime you start/ restart the service.
- Build the Project
go build ./cmd/api - Start API service to retrieve questions
- without apikey authentication
./api - with apikey authentication
./api -apikey=<your_api_key>
- without apikey authentication
- For more startup parameters
./api --help
- Setup migrate cli
- Setup Postgres and create a database
- Expose database service name
export GOQUIZ_DB=postgres://<username>:@localhost/<db_name>?sslmode=disable - Migrate the database, create necessary tables
migrate -path=./resources/migrations -database=$GOQUIZ_DB up - Build the Project
go build ./cmd/api - Scrap the questions and populate the database
./api -populate-db -db-dsn=$GOQUIZ_DB - Start API service to retrieve questions
- without apikey authentication
./api -db-dsn=$GOQUIZ_DB - with apikey authentication
./api -db-dsn=$GOQUIZ_DB -apikey=<your_api_key>
- without apikey authentication
- For more startup parameters
./api --help
- Get all categories
curl --request GET \ --url http://localhost:4000/v1/categories \ --header 'Authorization: Key 1234' - Get questions by category ID
curl --request GET \ --url 'http://localhost:4000/v1/questions?category_id=1' - Create new category
curl --request POST \ --url http://localhost:4000/v1/categories \ --header 'Content-Type: application/json' \ --data '{ "name":"Go Programming" }' - Create new question by category
curl --request POST \ --url http://localhost:4000/v1/questions \ --header 'Authorization: Key 1234' \ --header 'Content-Type: application/json' \ --data '{ "categ_id": 75, "text": "Which company created go programming language?", "answers": [ "Apple", "Google", "Amazon", "Facebook" ], "correct_answer": 2, "codeblock": "fmt.Println(\"Hello! Example codeblock\")", "explanation": "Go was originally designed at Google in 2007" }'
- Add GRPC Transport
- Add Structure logger
- Dockerize Deployment
- Deploy the API to Digital Ocean VPS
- Unit Testings for Model and APIs
- Add github CI to automate unit tests and deployment
- Update/ Delete endpoints for questions and categories
- Add web scraper to fetch MCQs from www.sanfoundry.com