7 releases

new 0.4.0-alpha.7 May 22, 2024
0.4.0-alpha.6 Mar 25, 2024
0.4.0-alpha.3 Jan 25, 2024

#247 in Database interfaces

Download history 76/week @ 2024-02-17 62/week @ 2024-02-24 111/week @ 2024-03-02 56/week @ 2024-03-09 49/week @ 2024-03-16 220/week @ 2024-03-23 47/week @ 2024-03-30 9/week @ 2024-04-06 97/week @ 2024-05-18

97 downloads per month

MIT/Apache

2MB
17K SLoC

Oxigraph CLI

Latest Version Crates.io downloads Conda actions status Gitter

Oxigraph CLI is a graph database implementing the SPARQL standard. It is packaged as a command-line tool allowing to manipulate RDF files, query them using SPARQL... It also allows to spawn a HTTP server on top of the database.

Oxigraph is in heavy development and SPARQL query evaluation has not been optimized yet.

Oxigraph provides different installation methods for Oxigraph CLI:

It is also usable as a Rust library and as a Python library.

Oxigraph implements the following specifications:

A preliminary benchmark is provided.

Note that Oxigraph CLI was previously named Oxigraph Server before version 0.4. Older versions are available under this name.

Packaging status

Installation

You need to have a recent stable version of Rust and Cargo installed.

To download, build and install the latest released version run cargo install oxigraph-cli. There is no need to clone the git repository.

To compile the command-line tool from source, clone this git repository including its submodules (git clone --recursive https://github.com/oxigraph/oxigraph.git), and execute cargo build --release in the cli directory to compile the full binary after having downloaded its dependencies. It will create a fat binary in target/release/oxigraph.

Some build options (cargo features) are available:

  • rocksdb-pkg-config: links against an already compiled rocksdb shared library found using pkg-config.
  • native-tls: Enables Oxigraph HTTP client for query federation using the host OS TLS stack (enabled by default).
  • rustls-native Enables Oxigraph HTTP client for query federation using Rustls and the native certificates.
  • rustls-webpki Enables Oxigraph HTTP client for query federation using Rustls and the Common CA Database certificates.

Usage

Run oxigraph serve --location my_data_storage_directory to start the server where my_data_storage_directory is the directory where you want Oxigraph data to be stored. It listens by default on localhost:7878.

The server provides an HTML UI, based on YASGUI, with a form to execute SPARQL requests.

It provides the following REST actions:

  • /query allows evaluating SPARQL queries against the server repository following the SPARQL 1.1 Protocol. For example:
    curl -X POST -H 'Content-Type:application/sparql-query' \
      --data 'SELECT * WHERE { ?s ?p ?o } LIMIT 10' http://localhost:7878/query
    
    This action supports content negotiation and could return Turtle, N-Triples, RDF/XML, SPARQL Query Results XML Format and SPARQL Query Results JSON Format.
  • /update allows to execute SPARQL updates against the server repository following the SPARQL 1.1 Protocol. For example:
    curl -X POST -H 'Content-Type: application/sparql-update' \
      --data 'DELETE WHERE { <http://example.com/s> ?p ?o }' http://localhost:7878/update
    
  • /store allows to retrieve and change the server content using the SPARQL 1.1 Graph Store HTTP Protocol. For example:
    curl -f -X POST -H 'Content-Type:application/n-triples' \
      -T MY_FILE.nt "http://localhost:7878/store?graph=http://example.com/g"
    
    will add the N-Triples file MY_FILE.nt to the server dataset inside of the http://example.com/g named graph. Turtle, N-Triples and RDF/XML are supported. It is also possible to POST, PUT and GET the complete RDF dataset on the server using RDF dataset formats (TriG and N-Quads) against the /store endpoint. For example:
    curl -f -X POST -H 'Content-Type:application/n-quads' \
      -T MY_FILE.nq http://localhost:7878/store
    
    will add the N-Quads file MY_FILE.nq to the server dataset.

Use oxigraph --help to see the possible options when starting the server.

It is also possible to load RDF data offline using bulk loading: oxigraph load --location my_data_storage_directory --file my_file.nq

Using a Docker image

Display the help menu

docker run --rm ghcr.io/oxigraph/oxigraph --help

Run the Webserver

Expose the server on port 7878 of the host machine, and save data on the local ./data folder

docker run --rm -v $PWD/data:/data -p 7878:7878 ghcr.io/oxigraph/oxigraph serve --location /data --bind 0.0.0.0:7878

You can then access it from your machine on port 7878:

# Open the GUI in a browser
firefox http://localhost:7878

# Post some data
curl http://localhost:7878/store?default -H 'Content-Type: text/turtle' -T ./data.ttl

# Make a query
curl -X POST -H 'Accept: application/sparql-results+json' -H 'Content-Type: application/sparql-query' --data 'SELECT * WHERE { ?s ?p ?o } LIMIT 10' http://localhost:7878/query

# Make an UPDATE
curl -X POST -H 'Content-Type: application/sparql-update' --data 'DELETE WHERE { <http://example.com/s> ?p ?o }' http://localhost:7878/update

Run the Web server with basic authentication

It can be useful to make Oxigraph SPARQL endpoint available publicly, with a layer of authentication on /update to be able to add data.

You can do so by using a nginx basic authentication in an additional docker container with docker-compose. First create a nginx.conf file:

daemon off;
events {
    worker_connections  1024;
}
http {
    server {
        server_name localhost;
        listen 7878;
        rewrite ^/(.*) /$1 break;
        proxy_ignore_client_abort on;
        proxy_set_header  X-Real-IP  $remote_addr;
        proxy_set_header  X-Forwarded-For $proxy_add_x_forwarded_for;
        proxy_set_header  Host $http_host;
        proxy_set_header Access-Control-Allow-Origin "*";
        location ~ ^(/|/query)$ {
            proxy_pass http://oxigraph:7878;
            proxy_pass_request_headers on;
        }
        location ~ ^(/update|/store)$ {
            auth_basic "Oxigraph Administrator's Area";
            auth_basic_user_file /etc/nginx/.htpasswd; 
            proxy_pass http://oxigraph:7878;
            proxy_pass_request_headers on;
        }
    }
}

Then a docker-compose.yml in the same folder, you can change the default user and password in the environment section:

version: "3"
services:
  oxigraph:
    image: ghcr.io/oxigraph/oxigraph:latest
    ## To build from local source code:
    # build:
    #   context: .
    #   dockerfile: server/Dockerfile
    volumes:
      - ./data:/data

  nginx-auth:
    image: nginx:1.21.4
    environment:
      - OXIGRAPH_USER=oxigraph
      - OXIGRAPH_PASSWORD=oxigraphy
    volumes:
      - ./nginx.conf:/etc/nginx/nginx.conf
      ## For multiple users: uncomment this line to mount a pre-generated .htpasswd 
      # - ./.htpasswd:/etc/nginx/.htpasswd
    ports:
      - "7878:7878"
    entrypoint: "bash -c 'echo -n $OXIGRAPH_USER: >> /etc/nginx/.htpasswd && echo $OXIGRAPH_PASSWORD | openssl passwd -stdin -apr1 >> /etc/nginx/.htpasswd && /docker-entrypoint.sh nginx'"

Once the docker-compose.yaml and nginx.conf are ready, start the Oxigraph server and nginx proxy for authentication on http://localhost:7878:

docker-compose up

Then it is possible to update the graph using basic authentication mechanisms. For example with curl: change $OXIGRAPH_USER and $OXIGRAPH_PASSWORD, or set them as environment variables, then run this command to insert a simple triple:

curl -X POST -u $OXIGRAPH_USER:$OXIGRAPH_PASSWORD -H 'Content-Type: application/sparql-update' --data 'INSERT DATA { <http://example.com/s> <http://example.com/p> <http://example.com/o> }' http://localhost:7878/update

In case you want to have multiple users, you can comment the entrypoint: line in the docker-compose.yml file, uncomment the .htpasswd volume, then generate each user in the .htpasswd file with this command:

htpasswd -Bbn $OXIGRAPH_USER $OXIGRAPH_PASSWORD >> .htpasswd

Build the image

You could easily build your own Docker image by cloning this repository with its submodules, and going to the root folder:

git clone --recursive https://github.com/oxigraph/oxigraph.git
cd oxigraph

Then run this command to build the image locally:

docker build -t ghcr.io/oxigraph/oxigraph -f server/Dockerfile .

Homebrew

Oxigraph maintains a Homebrew formula in a custom tap.

To install Oxigraph server using Homebrew do:

brew tap oxigraph/oxigraph
brew install oxigraph

It installs the oxigraph binary. See the usage documentation to know how to use it.

Systemd

It is possible to run Oxigraph in the background using systemd.

For that, you can use the following oxigraph.service file (it might be inserted into /etc/systemd/system/ or $HOME/.config/systemd/user):

[Unit]
Description=Oxigraph database server
After=network-online.target
Wants=network-online.target

[Service]
Type=notify
ExecStart=/PATH/TO/oxigraph serve --location /PATH/TO/OXIGRAPH/DATA

[Install]
WantedBy=multi-user.target

Migration guide

From 0.2 to 0.3

  • The cli API has been completely rewritten. To start the server run oxigraph serve --location MY_STORAGE instead of oxigraph --file MY_STORAGE.
  • Fast data bulk loading is not supported using oxigraph load --location MY_STORAGE --file MY_FILE. The file format is guessed from the extension (.nt, .ttl, .nq...).
  • RDF-star is now implemented.
  • All operations are now transactional using the "repeatable read" isolation level: the store only exposes changes that have been "committed" (i.e. no partial writes) and the exposed state does not change for the complete duration of a read operation (e.g. a SPARQL query) or a read/write operation (e.g. a SPARQL update).

Help

Feel free to use GitHub discussions or the Gitter chat to ask questions or talk about Oxigraph. Bug reports are also very welcome.

If you need advanced support or are willing to pay to get some extra features, feel free to reach out to Tpt.

License

This project is licensed under either of

at your option.

Contribution

Unless you explicitly state otherwise, any contribution intentionally submitted for inclusion in Oxigraph by you, as defined in the Apache-2.0 license, shall be dual licensed as above, without any additional terms or conditions.

Dependencies

~10–25MB
~419K SLoC