LOGRE (LOcal GRaph Editor)

Logre is an open-source UI that helps you visualize, edit, and explore RDF graph data through SPARQL endpoints.

Logre does not embed its own triple store. You can connect Logre to your own endpoint, or run it with the bundled RDF4J stack in containers.

Supported SPARQL endpoint technologies:

• Apache Jena Fuseki
• Eclipse RDF4J
• AllegroGraph
• Ontotext GraphDB

---

Stack overview (container mode)

Service	Purpose	Default port(s)
logre	Streamlit-based UI served by this repository	8501 (LOGRE_PORT)
rdf4j	Official rdf4j image (Server + Workbench)	8080 (RDF4J_SERVER_PORT)

Both services run under the dev Docker Compose profile and persist their state in volumes (logre_data, rdf4j_data).

---

Requirements

• A way to obtain the source code:
  • Git (recommended), or
  • Downloading the repository ZIP archive from GitHub.
• For make start: Linux/macOS shell (or WSL), make, Python 3 with venv, git, curl.
• For Docker/Podman mode: Docker Desktop or Docker Engine + Compose plugin (Podman + podman compose also works).
• For plain Python mode: Python 3.11+ and pip.

---

Run Logre (3 ways)

1) make start (historical workflow)

From repository root:

make start

This will switch to main, update from git, install dependencies in pipenv_logre, and run Streamlit.

Useful variants:

• make start-dev
• make reinstall
• make help

---

2) Docker / Podman (recommended for turnkey setup)

From repository root:

docker compose --profile dev up --build

Podman equivalent:

podman compose --profile dev up --build

If you need a full rebuild without cache:

docker compose --profile dev build --no-cache
docker compose --profile dev up

Open:

• Logre UI: http://localhost:8501 (or custom LOGRE_PORT)
• RDF4J Server API: http://localhost:8080/rdf4j-server
• RDF4J Workbench UI: http://localhost:8080/rdf4j-workbench/repositories
• Default repository endpoint: http://localhost:8080/rdf4j-server/repositories/logre

Optional port override:

cp .env.example .env

Then edit .env:

LOGRE_PORT=8502
RDF4J_SERVER_PORT=8081

RDF4J Workbench note:

Even if you expose RDF4J on another host port, Workbench connects to the internal URL http://rdf4j:8080/rdf4j-server. In a fresh browser session, enter this URL when prompted, then go to Repositories -> logre -> Use.

Common commands:

• Start detached: docker compose --profile dev up -d
• Stop stack: docker compose --profile dev down
• Tail logs: docker compose --profile dev logs -f logre
• Rebuild without cache: docker compose --profile dev build --no-cache
• Reset everything (delete persisted data): docker compose --profile dev down -v

---

3) Plain Python (manual local run)

From repository root:

python3 -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install -r requirements.txt
python -m streamlit run src/server.py

Then open http://localhost:8501.

When the Model page is opened, Logre auto-downloads src/lib/shacl-maker.js if it is missing.

---

Configuration & secrets

Logre stores its user configuration in a YAML file. If no config is present, Logre creates one from the bundled template.

Default locations (by OS):

• Linux: ~/.config/logre/logre-config.yaml
• macOS: ~/Library/Application Support/Logre/logre-config.yaml
• Windows: %APPDATA%\Logre\logre-config.yaml

Overrides (highest priority wins):

1. LOGRE_CONFIG_PATH (explicit file path)
2. LOGRE_CONFIG_HOME (directory override)
3. OS default locations above

Secrets should live in environment variables (or .env locally) and be referenced from the config with placeholders:

endpoints:
  - name: my-endpoint
    technology: RDF4J
    url: ${LOGRE_SPARQL_URL}
    username: ${LOGRE_SPARQL_USERNAME}
    password: ${LOGRE_SPARQL_PASSWORD}

Example .env (local, not committed):

LOGRE_SPARQL_URL=https://example.org/rdf4j-server/repositories/myrepo
LOGRE_SPARQL_USERNAME=admin
LOGRE_SPARQL_PASSWORD=secret

Docker specifics:

• LOGRE_CONFIG_PATH is set to /data/logre-config.yaml (persistent volume).
• The config is templated from docker/logre-config.yml on first run only.
• Data graph autoconfiguration runs automatically in Docker; in local runs it is opt-in via LOGRE_AUTOCONFIGURE_GRAPH=1.

Configuration migrations run automatically when the format changes.

---

License & documentation

Logre is released under the MIT License (see LICENSE).

• FAQ: documentation/faq.md
• Tests and QA notes: tests.md, things_to_test.md