A test home task as a part of the interview process to the company ciValue The current solution provides an POC version only.

Main used items:

• Java 8
• Spring Boot
• MongoDB
• Code-writting plugin: Lombok

The project has been configured to use MongoDB that is run in Docker (see ./docker-compose.yml) If you need to change connection parameters to MongoDB, please, edit the spring.data.mongodb.uri parameter in the src\main\resources\application.properties file

1. Run Docker containers

   docker compose -f "docker-compose.yml" up -d --build

2. Run service

   For now use your IDE for local running (Java 8 is required) Class path: src\main\java\com\civalue_test\Application.java

   !!! The service dockerization process is started but has not finished yet

   You can review commented part of the docker-compose.yml file

   and other documentation in the .info, .run, docker, settings folders to got main idea

Please, find below description of implemented APIs and technical details about used data models

route: /api/i_v1/product request body (JSON):

{
    "productId": "string",
    "category": "string",
    "brand": "string"
}

Example of usage:

curl --location --request POST 'localhost:3000/api/i_v1/product' \
    --header 'Content-Type: application/json' \
    --data-raw '{
        "productId": "ID4",
        "category": "TEST_CATEGORY4",
        "brand": "TEST_BRAND4"
    }'

route: /api/i_v1/personalized_list request body (JSON):

{
    "shopperId": "string",
    "shelf": [
        {
            "productId": "string",
            "relevancyScore": double
        }
    ]
}

Example of usage:

curl --location --request POST 'localhost:3000/api/i_v1/personalized_list' \
    --header 'Content-Type: application/json' \
    --data-raw '{
        "shopperId": "SID2",
        "shelf":[
            {
                "productId": "ID2",
                "relevancyScore": 0.1
            },
            {
                "productId": "ID3",
                "relevancyScore": 0.2
            },
            {
                "productId": "ID4",
                "relevancyScore": 0.3
            }
        ]
    }'

Supported query parameters:

• shopperId - required, could not be bblank
• category - optional, by default is empty, if blank -ignored
• brand - optional, by default is empty, if blank -ignored
• limit - optional, by default has value 10

Example of usage:

curl --location --request GET 'localhost:3000/api/ec_v1/product?shopperId=SID1&category=&brand&limit='

Supported query parameters:

• productId - required, could not be bblank
• limit - optional, by default has value 10

Example of usage:

curl --location --request GET 'localhost:3000/api/ec_v1/shopper?productId=ID4&limit='

Database schema:

collection of documents with the name "personalized_list" that have the following structure

{
    "shopperId": "not empty string, shoper's unique ID, e.g. 'S-1000'",
    "shelf": [ // array of embedded documents
        {
            "productId": "string, product's unique ID, e.g. 'MB-2093193398'",
            "relevancyScore": "double, special score value, e.g.31.089209569320897"
        }
        ...
    ]
}

collection of documents with the name "product" that have the following structure

{
    "productId": "string, product's unique ID, e.g. BB-2144746855",
    "category": "string, custom category name, e.g. 'Babies'",
    "brand": "string, custom brand name, e.g. 'Babyom'"
}

• Finish dockerization of main service
• Data consistency (the Shooper entity should be introduced)
• Caching for eCommerce part of APIs (it require definition of caching keys and evicting policy)
• Review necessity of splitting internal and external APIs into two independent services (as they have different requirements regarding performance and security)
• Introducing logging (local loging and sending to a company monitoring layer)
• Introducing metric collection (sending to a company monitoring layer)