q-server

Test server for upload bulk data

Configuration

To configure server was used environment approach.

• APP_SQLITE_DB_PATH defines path to database. REQUIRED
• APP_SERVER_PORT defines port on which server will listen. Default is 8080.
• APP_LOG_LEVEL is used to define log level. Possible values are debug, info, warn, error, fatal.Default is info.
• APP_SQLITE_DB_POOL_SIZE defines pool size of connections to database.
• APP_REQUESTS_LIMIT defines request limit to server. Default is 100.

Start application

There are some ways to launch server

1. Run standalone application

APP_SQLITE_DB_PATH=assets/schema.sqlite go run main.go

2. Run docker

docker build -t q-server .
docker run -it --rm -p 8080:8080 -v $(pwd)/assets:/app/data --env  APP_SQLITE_DB_PATH=/app/data/schema.sqlite q-server

3. Run docker-compose

docker-compose up

Usage

There are two endpoints to upload credit transfers:

• /transfer

curl --request POST 'http://127.0.0.1:8080/transfer' \
--header 'Content-Type: application/json' \
--data-raw '{
    "organization_name": "ACME Corp",
    "organization_bic": "OIVUSCLQXXX",
    "organization_iban": "FR10474608000002006107XXXXX",
    "credit_transfers": [
        {
            "amount": "23.17",
            "currency": "EUR",
            "counterparty_name": "Bip Bip",
            "counterparty_bic": "CRLYFRPPTOU",
            "counterparty_iban": "EE383680981021245685",
            "description": "Neverland/6318"
        },
        {
            "amount": "200",
            "currency": "EUR",
            "counterparty_name": "Daffy Duck",
            "counterparty_bic": "DDFCNLAM",
            "counterparty_iban": "NL24ABNA5055036109",
            "description": "2020/RabbitSeason/"
        }
    ]
}'

• /upload

curl --request POST 'http://127.0.0.1:8080/upload' \
--header 'Content-Type: multipart/form-data' \
--form 'file=@"/PATH/TO/transfers.json"'

TODO

This approach will be work fine with count transfers less than 1000. If there are expecting more transfers in one request, should split INSERT INTO queries to bulks.

Store amounts in cents is not good approach when we need to get some statistic (average, percentage etc). It's better to store cents which multiplied to 1000000 (common practice).