WARNING: The contents of this repository are old and we aren't supporting it any more. The examples probably don't work with the latest versions of BigchainDB. We're keeping this repository around (in read-only mode) because it explored many interesting ideas.
Tutorials, Examples and Experiments with BigchainDB. Welcome to the BigchainDB application laboratory!
Kyber is a full suite of BigchainDB repo's including:
- BigchainDB server
- BigchainDB client Tutorials ( python, JavaScript)
- Example applications
- Experimental stuff
All versions of the above:
- Are in sync with the master branch of each BigchainDB repo
- Might have experimental features (watch out that you don't burn yourself ;-) )
Getting started with BigchainDB? Have a look at our docs:
- the HTTP API
- a Python driver
- a (minimal) JavaScript driver for creating transactions
First things first. You'll need a BigchainDB server to get going with the API. If you want run the server locally follow these steps:
Clone this repo (using submodules)
git clone git@github.com:bigchaindb/kyber.git --recursive and
cd kyberYou must have docker, docker-compose (and make) installed.
These versions or higher should work:
docker:v1.13.0docker-compose:v1.7.1
To spin up the services, simple run the make command, which will orchestrate docker-compose
makeThis might take a few minutes, perfect moment for a ☕!
Once docker-composed has built and launched all services, have a look:
docker-compose ps Name Command State Ports
----------------------------------------------------------------------------------------------------------------
kyber_bdb-server_1 bigchaindb start Up 0.0.0.0:49984->9984/tcp
kyber_rdb_1 rethinkdb --bind all Up 0.0.0.0:32772->28015/tcp, 29015/tcp, 0.0.0.0:58585->8080/tcp
...
Which means that the internal docker port for the API is 9984
and the external one is 49984.
The external ports might change, so for the following use the ports as indicated by docker-compose ps.
You can simply check if it's running by going to http://localhost<external-docker-port-bdb-server>.
Also you can access the RethinkDB dashboard on http://localhost:<external-docker-port-rdb>, which is 58585 in our case
If you already built the images and want to restart:
make restartStop the containers with
make stopNo make? Launch the services manually:
Launch RethinkDB:
docker-compose up -d rdbWait about 10 seconds and then launch the server:
docker-compose up -d bdb-serverOk, so this goes well under LINUX, as most python devs know. OSX might be fine too, see the BigchainDB driver docs
If you want to run the Python tutorial with the experimental driver,
you'll need to install the local python driver and server of bigchaindb
(in a python3 virtual environment would be smart e.g. virtualenv venv -p python3 && source venv/bin/activate):
cd drivers/python/
pip install -e .
cd ../../Here is a list of Python tutorials for a BigchainDB client:
- simple_transactions.py:
Prepare, sign and post basic
CREATE,TRANSFERtransactions - assets_unspents.py: Create, transfer and list assets, unspents, etc.
- divisible_transactions.py: Split and combine transactions with divisible assets
- cryptoconditions_transactions.py: Create custom UTXO scripts using py-crypto-conditions
Descend into the tutorials directory
cd tutorials/The tutorials require a BDB_SERVER_URL.
If you are running the server locally with docker we needed to remember the external port <external-docker-port> of the API in docker (run docker-compose ps in the repo root).
In our case the BDB_SERVER_URL was http://localhost:49984.
For example:
PYTHONPATH=. BDB_SERVER_URL=<bigchaindb-server-url> python 01_simple_transactions/simple_transactions.pyIf you want to run the JavaScript tutorial, you'll need node and npm:
These versions or higher should work:
node:v6.2.2npm:v3.9.5
Here is a list of JavaScript tutorials:
- simple_transactions.js:
Prepare, sign and post basic
CREATE,TRANSFERtransactions - assets_unspents.js: Create, transfer and list assets, unspents, etc.
- divisible_transactions.js: Split and combine transactions with divisible assets
- [TODO] Create custom UTXO scripts using js-crypto-conditions
The JavaScript driver will be served from the local repo (until stable enough to appear on npm.registry)
This means we have to link js-bigchaindb-quickstart from the local drivers/javascriptrepo before install.
Here is how you do that:
cd drivers/javascript
npm install
npm link
cd ../../tutorials
npm link js-bigchaindb-quickstartWe can now build the JavaScript bundles using npm.
Make sure you are in the tutorials directory first.
The tutorials require a BDB_SERVER_URL. If you are running the server locally with docker we needed to remember the external port <external-docker-port> of the API in docker (run docker-compose ps in the repo root).
In our case the BDB_SERVER_URL was http://localhost:49984.
BDB_SERVER_URL=<bigchaindb_server_url> npm installA few more sips of ☕ later...
If all goes well, you'll see webpack spitting out the bundles. That's a good sign!
Once we have the bundles under tutorials/build, we're all set to inject them into a html file.
There are some simple webpages in each tutorial that can be served under the tutorials directory.
Due to same-origin policy, you'll need to start a local webserver (in another terminal) to serve the folder tutorials.
Here are some go-to's to server local html files:
python:python -m SimpleHTTPServerpython3:python3 -m http.servernode: thelocal-web-servermodule see https://www.npmjs.com/package/local-web-server
For example, using the python server uses port:8000.
You can access a (blank) page that will run the JavaScript in the background:
- http://localhost:8000/01_simple_transactions/simple_transactions.html
- http://localhost:8000/02_assets_unspents/assets_unspents.html
- http://localhost:8000/03_divisible_transactions/divisible_transactions.html
The tutorials only use JavaScript without frontend visuals. (sorry for that, could go into the ideabox)
However statements are printed in the dev console (F12 in the browser, and F5 to reload)
Launch webpack in watch mode and wait for the bundles to be emitted:
BDB_SERVER_URL=<bigchaindb_server_url> webpack -w As long as the webpack watcher is running, every code change will be trigger a new build of the affected bundle(s).
If you used the make approach, then examples should be running under the Docker container examples-client-frontend:
Name Command State Ports
---------------------------------------------------------------------------------------------------------
kyber_examples-client- node server.demo.js Up 0.0.0.0:33000->3000/tcp
frontend_1 Typically the port is 33000, so you can simple see the examples on http://localhost:33000/.