Weather Tool

Tool for downloading historical and real-time data from 33 weather stations in Latvia. It aggregates and displays data for different time periods, cities, and weather parameters.

Setup

Rename .env-sample to .env and fill in credentials

Run

docker-compose up --build --no-cache --force-recreate

Local synthetic weather data

The local database can be populated with deterministic hourly sample data for all 33 stations. The seed covers the previous 18 months through the current hour and can be run repeatedly without creating duplicates.

docker compose --profile tools run --rm seed

This workflow is intended only for development and UI testing. It uses the normal weather table and application APIs; it does not provide LVGMC forecast CSV or HARMONIE GRIB fixtures.

Web

http://0.0.0.0:9090/

nginx proxy config

server {
    listen 80;
    server_name laikazinas.lsm.lv;

    location / {
        proxy_pass http://localhost:9090; # Forward requests to the Scala app
        proxy_http_version 1.1;
        proxy_set_header Upgrade $http_upgrade;
        proxy_set_header Connection 'upgrade';
        proxy_set_header Host $host;
        proxy_cache_bypass $http_upgrade;
    }
}

Local development

// 1st terminal
podman-compose up postgres

// 2nd terminal
sbt run

// 3rd terminal
cd web/
npm run dev

Run one file sbt "runMain grib.GribParserTest"

Backend Tech:

• scala: cats-effects, http4s, fs2, circe, scalatest
• postgres

Frontend Tech:

• SolidJS
• Vite

Fly commands

// suspend instance
fly scale count 0

// check display
fly ssh console
df -h

fly scale memory 512