The ib_insync
package is build on top of the Python API
from Interactive Brokers. The objective is to make it as
easy as possible to use the API to its fullest extent.
The main features are:
- An
IB
component that automatically keeps in sync; - An easy to use linear style of programming (no more callbacks);
- A fully asynchonous framework based on asyncio for advanced users;
- Interactive operation with live data in Jupyter notebooks.
pip3 install -U ib_insync
Requirements:
- Python version 3.6 or higher;
- The Interactive Brokers Python API version 9.73.06 or higher;
- A running TWS or IB gateway application (version 969 or higher). Make sure the API port is enabled and 'Download open orders on connection' is checked.
To install packages needed for the examples and notebooks:
pip3 install -U jupyter numpy pandas
This is a complete script to download historical data:
from ib_insync import *
ib = IB()
ib.connect('127.0.0.1', 7497, clientId=1)
contract = Forex('EURUSD')
bars = ib.reqHistoricalData(contract, endDateTime='', durationStr='30 D',
barSizeSetting='1 hour', whatToShow='MIDPOINT', useRTH=True)
# convert to pandas dataframe:
df = util.df(bars)
print(df[['date', 'open', 'high', 'low', 'close']])
Output:
date open high low close 0 2017-08-13 23:15:00 1.182850 1.183100 1.182100 1.182400 1 2017-08-14 00:00:00 1.182400 1.182450 1.181875 1.182175 2 2017-08-14 01:00:00 1.182175 1.182675 1.181900 1.182525 ... 719 2017-09-22 22:00:00 1.194425 1.195425 1.194225 1.195050
Be sure to take a look at the notebooks and the recipes too.
The insync user group is the place to discuss IB-insync and anything related to it.
The software is provided on the conditions of the simplified BSD license.
This project is not affiliated with Interactive Brokers Group, Inc.'s.
- historical ticks and realtime bars now return time in UTC
- IBController added
- openOrder callback added
- default arguments for ib.connect() and ib.reqMktData()
- minimum API version is v9.73.06
- tickByTick support
- automatic request throttling
- ib.accountValues() now works for multiple accounts
- AccountValue.modelCode added
- Ticker.rtVolume added
- workaround for IBAPI v9.73.06 for Contract.lastTradeDateOrContractMonth format
- util.tree() method added
error
callback signature changed to (reqId, errorCode, errorString)accountValue
andaccountSummary
callbacks added
- util.useQt fixed for use with Windows
- Fix for ib.schedule()
- Import order conditions into ib_insync namespace
- util.useQtAlt() added for using nested event loops on Windows with Qt
- ib.schedule() added
- Fixed conditional orders
- FlexReport added
- Fixed issue #22
- Ticker.vwap field added (for use with generic tick 233)
- Client with master clientId can now monitor orders and trades of other clients
barUpdate
event now used also for reqRealTimeBars responsesreqRealTimeBars
will return RealTimeBarList instead of list- realtime bars example added to bar data notebook
- fixed event handling bug in Wrapper.execDetails
- BarDataList now used with reqHistoricalData; it also stores the request parameters
- updated the typing annotations
- added
barUpdate
event toIB
- bar- and tick-data notebooks updated to use callbacks for realtime data
- ticker.marketPrice adjusted to ignore price of -1
- ticker.avVolume handling fixed
- realtimeBar wrapper fix
- context manager for IB and IB.connect()
- compatibility with upcoming ibapi changes
- added
error
event toIB
- notebooks updated to use
loopUntil
- small fixes and performance improvements
- new IB.reqHistoricalTicks API method
- new IB.loopUntil method
- fixed issues #4, #6, #7
- fixed swapped ticker.putOpenInterest vs ticker.callOpenInterest
- fixed wrapper.tickSize regression
- support for realtime bars and keepUpToDate for historical bars
- added option greeks to Ticker
- new IB.waitUntil and IB.timeRange scheduling methods
- notebooks no longer depend on PyQt5 for live updates
- notebooks can be run in one go ('run all')
- tick handling bypasses ibapi decoder for more efficiency
- IB.whatIfOrder() added
- Added detection and warning about common setup problems
- Removed import from ipykernel
- Removed dependencies for installing via pip
- added lots of request methods
- order book (DOM) added
- notebooks updated
- Added UTC timezone to some timestamps
- Fixed issue #1
- Initial release
Good luck and enjoy,
author: | Ewald de Wit <[email protected]> |
---|