The OANDA Trading Environment is built using the OANDA REST-API by making use of the https://github.com/oanda/oandapy API-wrapper.
Main part is the OANDAd daemon that parses the streaming quotes in configurable timeframes, by 1 minute, 5 minutes, 15 minutes etc. This makes it produce streaming candles.
As of version 0.3.0 OANDAd detects automatically the use of REST-V1 / REST-V20, by the account_id string in the OANDAd.cfg.
Candle data:
{"data": {"instrument": "EUR_JPY",
"granularity" : "M1",
"start": "2015-09-02 15:36:00"
"end": "2015-09-02 15:37:00",
"completed": True,
"data": {"high": 134.967,
"open": 134.962,
"last": 134.9565,
"low": 134.9475,
"volume": 19
},
}
}
The larger timeframes can be requested using the API.
Streaming data can be controlled by the 'fabricate' setting in the 'streamer:' config section.
- atEndOfTimeFrame - the default mode is to fabricate completed timeframes
- dancingBear - generates dancing bear records with the 'completed' : False, until the record is completed. At that moment the record gets the status 'completed' : True.
- dancingBearHighLowExtreme - generate dancing bear records only when the high or the low changes
The dancingBear setting generates as many records as ticks are received. This can be a lot. A compromise is the dancingBearHighLowExtreme. In case of extreme market moves the records will be generated also, but in a less volatile market less records will be generated.
When a timeframe is completed it can be handled by one or more plugins. Plugins have a configfile based on the name of the plugin-file, but in lowercase. The plugins can be found under etc/OANDA/plugins and the plugin configs under etc/OANDA/config/plugins.
Plugins need to be enabled in the config file, see the 'plugins:' section in etc/OANDA/config/OANDAd.cfg.
The environment comes with a few plugins:
This plugin can be configured to 'publish' the candle using a publisher/subscriber mechanism. This is achieved by using the 0MQ library and the python binding for it: pyzmq
Other trading applications can easily subscribe to receive the candle data. See here for a ZMQ subscription example.
This plugin is enabled by default.
This plugin can be configured to write candle records to a flatfile in a directory structure.
Example:
/tmp/oandadb
|-- BCO_USD
| |-- M1
| | `-- cache
| |-- M15
| | `-- cache
| `-- M5
| `-- cache
|-- DE30_EUR
| |-- M1
| | `-- cache
| |-- M15
| | `-- cache
| `-- M5
| `-- cache
|-- EUR_CHF
| |-- M1
| | `-- cache
| |-- M15
| | `-- cache
| `-- M5
| `-- cache
|-- EUR_GBP
| |-- M1
| | `-- cache
| |-- M15
| | `-- cache
| `-- M5
| `-- cache
|-- EUR_JPY
| |-- M1
| | `-- cache
| |-- M15
| | `-- cache
| `-- M5
| `-- cache
|-- EUR_USD
|-- M1
| `-- cache
|-- M15
| `-- cache
`-- M5
`-- cache
The MySQL plugin can be configured to insert records into a database. This plugin is provided as an example, since it needs details that depend on your databasemodel.
The desired approach is to create stand-alone applications that subscribe for quotes, see, example for a ZMQ subscription example.
Though it is possible to use the plugin facility to perform auto-trading, the way to go is to use ZMQ client and subscribe for quotes. This way you can completely isolate your trading code from the OANDAd daemon.
The enviroment makes use of a token that gives access to crucial information.
Please pay attention to where you install this software. Never use this on a system that is not owned by you.
Make sure to secure your system as much as possible:
- restrict network access
- make no use of, or limit other network services (NFS, SAMBA, printserver etc.)
- limit user access, preferable only you
- use encryption
- who about physical access ?
To access the OANDA services you will need a token, see https://developer.oanda.com for details.
Install by using a virtual environment and git:
$ cd <somewhere>
$ mkdir OANDA
$ cd OANDA
$ virtualenv venv
Optionally use --system-site-packages to use the standard available packages for the python modules available on your system: pyyaml, pyzmq. Check for the packages on the distribution you use.
$ . ./venv/bin/activate
$ git clone https://github.com/hootnot/oanda-trading-environment.git
$ cd oanda-trading-environment
$ python setup.py install
OANDA has not made the oandapy module pip installable. A hack to get oandapy installed as a standalone module:
$ pip install git+https://github.com/hootnot/oandapy
This will install the latest oandapy using the setup.py from the oandapy fork.
$ pip list | grep oanda
oanda-trading-environment (0.0.1)
oandapy (0.1)
Install from pypi:
using a virtual environment:
$ cd <somewhere>
$ mkdir OANDA
$ cd OANDA
$ virtualenv [--system-site-packages] venv
$ . ./venv/bin/activate
$ pip install git+https://github.com/hootnot/oandapy
$ pip install --no-binary :all: oanda-trading-environment
using a system install:
$ sudo pip install oanda-trading-environment
$ sudo pip install git+https://github.com/hootnot/oandapy
Configure the OANDAd.cfg config file and start the daemon.
After installing you need to configure the environment by editing the config file etc/OANDA/config/OANDAd.cfg. This is a YAML based configfile.
Configure the environment, default set to practice and the token. Alter the list of instruments you want to follow.
Though accounts can be requested using the API, the environment uses the account setting primarily to initialize the OANDAd for the streaming quotes. Therefore you need to configure the account_id also.
The pubsub plugin publishes by default at localhost, port 5550. These can be altered in the 'pubsub' config: etc/OANDA/config/plugins/pubsub.cfg.
OANDAd is built using daemoncle. The 'start', 'status' and 'stop' commands are implemented. The daemon forks itself and the child will process the stream. When there are issues, TIME-OUT for instance, the child will exit and a new child will be spawned.
$ OANDAd start
Starting OANDAd ... OK
$ OANDAd status
OANDAd -- pid: 51931, status: sleeping, uptime: 0m, %cpu: 0.0, %mem: 1.8
$ OANDAd stop
Stopping OANDAd ... OK
The daemon will process streaming quotes now and process the timeframes as configured. Timeframes are currently based on the midprice of bid/ask.
The ticks received from the stream are written to a logfile:
streamdata.<date>
The daemon itself logs to OANDAd.log
Loglevel and the streamdata logfile extension is configurable. Check the OANDAd.cfg file for details.
This simple piece of code acts as a subscriber to the daemon. All completed timeframes are written to stdout. Using ZMQ make it easy to program different strategies completely independent from each other. By using 'topics' it is possible to subscribe for a certain time granularity like M1, M5 etc. Check the ZMQ for details.
import zmq
context = zmq.Context()
socket = context.socket( zmq.SUB)
socket.connect("tcp://127.0.0.1:5550")
socket.setsockopt(zmq.SUBSCRIBE, "")
socket.setsockopt( zmq.RCVBUF, 1000)
while True:
msg = socket.recv()
print "GOT: ", msg
This will show candle data like below, every time a timeframe is completed.
GOT: {"data": {
"instrument": "EUR_GBP",
"granularity": "M1",
"start": "2015-09-04 17:45:00",
"end": "2015-09-04 17:46:00",
"completed": True,
"data": {
"high": 0.734445,
"open": 0.734399,
"last": 0.73437,
"low": 0.734345,
"volume": 16
}
}
}
GOT: {"data": {
"instrument": "EUR_JPY",
"granularity": "M1",
"start": "2015-09-04 17:45:00",
"end": "2015-09-04 17:46:00",
"completed": True,
"data": {
"high": 132.629,
"open": 132.619,
"last": 132.6185,
"low": 132.608,
"volume": 15
}
}
}
GOT: {"data": {
"instrument": "SPX500_USD",
"granularity": "M1",
"start": "2015-09-04 17:45:00",
"end": "2015-09-04 17:46:00",
"completed": True,
"data": {
"high": 1915.35,
"open": 1914.75,
"last": 1915.25,
"low": 1914.75,
"volume": 33
}
}
}
The oanda-trading-environment software is a personal project. I have no prior or existing relationship with OANDA.
If you have any questions regarding this software, please take a look at the documentation first.
If you still have questions/issues you can open an issue on Gitub: https://github.com/hootnot/oanda-trading-environment