Usage documentation is located at https://docs.flur.ee.
Fluree is an immutable, temporal, ledger-backed semantic graph database that has a cloud-native architecture.
This repository is a stateless database as a library and designed to be utilized in conjunction with the
Fluree Ledger which maintains all state. This database
can be run in containers and dynamically scale to any desired load, can be embedded inside
of your applications (Clojure, NodeJS for now) or can run as a stand-alone JVM service.
This database can also be built as a web-worker, and be embedded inside the browser. Thus
far, a React Wrapper (Beta) has been developed that allows
you to create real-time apps by wrapping your React components with queries (GraphQL or FlureeQL).
It is also possible to run Fluree in a "serverless" manner, where by utilizing Fluree SmartFunctions
to embed data security along side your data (Data Defending Itself), you can have a permissioned
application with just a single-page application (i.e. React) and Fluree Ledgers, but no application server.
Fluree includes time travel, allowing you to instantly query as of any historical moment in time,
and even allows the abilty to stage proposed transactions to time travel into the future, to a hypothesized version
of your data.
The best way to get started with Fluree is to go to the Getting Started page
All contributors must complete a Contributor License Agreement.
- Install clojure tools-deps (version 126.96.36.1993 or later).
brew install clojure/tools/clojure
- Arch Linux:
pacman -S clojure
- Install Node & NPM
brew install node
- Arch Linux:
pacman -S nodejs
- Install babashka
brew install borkdude/brew/babashka
make -j to run tasks in parallel.
make deps - install all local dependencies
make - make everything below
make jar - make Java JAR file
make install - install jar file into local .m2/maven
make clean - clean all build directories/files
make test - run all automated tests below
make cljtest - run all CLJ tests
make cljstest - run CLJS tests in headless Chrome & NodeJS
make cljs-browser-test - run CLJS tests in headless Chrome
- Needs karma-cli installed globally:
npm install -g karma-cli
- Needs Google Chrome browser installed
make cljs-node-test - run CLJS tests in NodeJS
make nodejs-test - run node package in node
make browser-test - run browser package in headless Chrome
This applies to CLJ tests only, not CLJS.
clojure -X:cljtest :kaocha.filter/focus [focus-spec]
focus-spec can be a test namespace or a fully-qualified
var. Note that the square brackets around the
focus-spec must be present in
the command, they are NOT there to indicate "optional" or "placeholder" in the
This feature comes from the test runner kaocha which has
In order to get a Node or web browser CLJS REPL running, you need to do the
npx shadow-cljs watch flureenjs (or
flureedb for a browser REPL)
- (Node only) Run
node out/nodejs/flureenjs.js in a separate shell
- (browser only) Connect to
http://localhost:9630/ in your browser
- This seems currently broken though. It first gets stuck at
"shadow-cljs Loading..." with an "unknown route" error in the JS console.
If you reload that seems to resolve. But even once the dashboard loads
and you connect a REPL, it says "No available JS runtime."
- Connect an nREPL to the port specified in
- Inside that REPL run
(shadow/repl :flureenjs) (or
:flureedb for a browser
- Try running something CLJS-specific to ensure you've got a working CLJS REPL
- For example: