This is a simple python / django application that utilises the smileycoin-cli interface to populate a relational database with the block chain information. A simple Twitter client is also built in.
The database can be queried using the django ORM or a web based API.
You probably need to set the directory of smileycoin-cli in the settings file ./smileychain/local_settings.py, as with the database described below, to SMILEYCOIN_CLI_LOCATION. It defaults to /usr/local/bin/.
The project is a final project in the course Crypotcurrencies STÆ532M2019H. Work is done by ves4 and ava7.
- To create an infrastructure to support quick development on top of the SmileyCoin blockchain, querying a relational database for analytics or information is much more efficient than targeting the blockchain directly.
- To create a Twitter bot that scans OP_RETURN values on transactions and posts them, maintaining a history of already posted messages and messages that do not nicely convert into ASCII.
You need to have a local smileycoind server running and smileycoin-cli setup, see https://tutor-web.info/smileycoin/download.
If you are starting it up for the first time make sure the configuration has txindex=1 set.
Setup an empty database that you can write to and configure smileychain/local_settings.py with the connection information.
Use a virtual environment or the like to install the packages in requirements.txt with pip install -r requirements.txt.
Migrate the database with python manage.py migrate, this set's up relevant tables and modifications.
Note: You may be able use a file based sqlite database but it is not recomended for anything but simple testing.
To populate the database with the blockchain information simply run
python manage.py scrape FROM_BLOCK TO_BLOCK
Where FROM_BLOCK and TO_BLOCK are block indexes.
You can use smileycoin-cli getblockchaininfo to fetch the highest available block number.
Simply run python manage.py runserver and open up the website mentioned in your browser (default http://127.0.0.1:8000/). This exposes a web interface for the REST api which enables navigation of some endpoints corresponding to tables in the database or models accesible through the Django ORM.
{
"blocks": "http://127.0.0.1:8000/blocks/",
"transactions": "http://127.0.0.1:8000/transactions/",
"vouts": "http://127.0.0.1:8000/vouts/",
"vins": "http://127.0.0.1:8000/vins/",
"addresses": "http://127.0.0.1:8000/addresses/",
"op_returns": "http://127.0.0.1:8000/op_returns/"
}Some of these have filter and search options built in that can be easily extended or configured as per the Django REST Framework documentation at https://www.django-rest-framework.org (see ./block/api/views.py).
Run python manage.py shell or python manage.py shell_plus to open up the Django shell with models and modules loaded for direct and efficient interaction with the ORM and db. E.g.
In [1]: OpReturn.objects.filter(message__icontains="hello").count()
Out[1]: 23
In [2]: OpReturn.objects.filter(message__icontains="hello").last().message
Out[2]: 'hello :)'A production ready configuration can be setup with e.g. nginx and gunicorn. You might want to do this with a docker-compose file using a pre built database/image which may be suplied at a later time.
Some simple functionality to extract OP_RETURN messages and post to Twitter (using tweepy) is included. See the the twitter folder for relevant code.
Use the management command python manage.py op_return to scrape interesting op return messages from the chain. These
can be viewed at something similar to http://127.0.0.1:8000/op_returns/.
To set up a twitter bot, first create a user and a new app with the keys needed. Then in the settings file or your local_settings populate
TWITTER_API_KEY = ""
TWITTER_API_SECRET = ""
TWITTER_ACCESS_TOKEN = ""
TWITTER_ACCESS_TOKEN_SECRET = ""
To post new OP_RETURN messages to your account simply run
python manage.py update_twitter
To see how it can look checkout https://twitter.com/return_op
To update the database, parse incoming blocks and push to twitter, run (see ./twitter/management/command/check_and_update.py)
python manage.py check_and_update
This is a good candidate for something to run in cron. To run every 10 minutes, run crontab -e and add:
*/10 * * * * /home/user/.virtualenvs/smiley/bin/python /home/user/SmileycoinAPIBot/smileychain/manage.py check_and_update >& /tmp/smly.log
For educational purposes it might be of interest to setup a fully functioning web based
command line interface against the database or local SmileyCoin server using the same syntax and appearance as smilecoin-cli, as long as input is sanitized and wallets individualized.
Currently, the implementation shells out to smileycoin-cli, ideally this should be implemented as rpc calls directly to the daemon.