AlertaDengueAnalise

This repository runs the Alerta Dengue analytics pipeline using R within a Conda environment, orchestrated by Makim tasks.

It is designed to run against a PostgreSQL database (local container or remote), generate per-state .RData artifacts, and produce SQL update scripts for downstream systems.

Requirements

• Linux
• Conda distribution (recommended: Miniforge/Mambaforge)
• mamba (recommended) or conda
• PostgreSQL database available (e.g., Docker container)

Quick start

1) Create the Conda environment

Create the environment from the repository conda spec:

mamba env create -f conda/base.yaml -n alertadengueanalise -y
conda activate alertadengueanalise

2) Install R dependencies (CRAN + GitHub)

makim deps.install --env alertadengueanalise

Optional smoke test:

makim deps.check --env alertadengueanalise

3) Configure database access

Create a .env file at the repo root with your Postgres connection settings.

Supported variables:

• DB_HOST
• DB_PORT
• DB_NAME
• DB_USER
• DB_PASSWORD
• DB_SSLMODE (optional)

Example:

ALERTA_DATA_RELATORIO=${ALERTA_DATA_RELATORIO}
ALERTA_OUT_DIR=${ALERTA_OUT_DIR}
ALERTA_DB_HOST=${ALERTA_DB_HOST}
ALERTA_DB_PORT=${ALERTA_DB_PORT}
ALERTA_DB_NAME=${ALERTA_DB_NAME}
ALERTA_DB_USER=${ALERTA_DB_USER}
ALERTA_DB_PASSWORD=${ALERTA_DB_PASSWORD}
ALERTA_DO_SCP=${ALERTA_DO_SCP} 
ALERTA_SCP_ENDPOINT=${ALERTA_SCP_ENDPOINT}
ALERTA_SCP_PATH=${ALERTA_SCP_PATH}

Check connectivity:

makim db.check --env alertadengueanalise

4) Run the BR pipeline for a given epidemiological week

Run the pipeline for week YYYYWW (example: 202601):

makim pipeline.run-br --env alertadengueanalise --week 202601

Documentation

• Pipeline behavior and steps: docs/WORKFLOW.md
• Outputs and where to find them: docs/OUTPUTS.md

Troubleshooting

• If you see missing R packages during execution, run:

  makim deps.install --env alertadengueanalise

• If the pipeline runs long with little console output, check:

  • CPU usage: ps -o pid,etime,pcpu,pmem,cmd -p <PID>
  • New files created under main/alertas/<YYYYWW>/
  • Log file under logs/ (if enabled by your pipeline wrapper)

License