Skip to content

oliviermilla/InteractiveBrokers.jl

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

99 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

InteractiveBrokers

Build Status codecov

A Julia implementation of Interactive Brokers API

InteractiveBrokers is a native Julia client that implements Interactive Brokers API to communicate with TWS or IBGateway.

It aims to be feature complete, however it does not support legacy versions. Currently, only API versions v176+ are supported.

The package design follows the official C++/Java IB API, which is based on an asynchronous communication model over TCP.

The package was first developed by lbilli in Jib.jl. lbilli still maintains his package and APIs and functionalities are integrated into InteractiveBrokers.jl.

What are the differences with Jib.jl then?

  1. InteractiveBrokers.jl is published to Julia's General Repository.
  2. DataFrame integration is done through an extension, allowing to keep the same API as Jib.jl but with lighter dependencies if you don't need DataFrames per se.
  3. User-provided callbacks can have an optional object upon which to dispatch. (see example below)

The development of InteractiveBrokers.jl was first motivated to integrate InteractiveBrokers with Lucky.jl trading framework.

Installation

To install from Julia General Repository:

] add InteractiveBrokers.jl

To install from GitHub:

] add https://github.com/oliviermilla/InteractiveBrokers.jl

Usage

The user interacts mainly with these two objects:

  • Connection: a handle holding a connection to the server
  • Wrapper: a container for the callbacks that are invoked when the server responses are processed.

Other data structures, such as Contract and Order, are implemented as Julia struct and mirror the respective classes in the official IB API.

A complete minimal working example is shown. For this code to work, an instance of IB TWS or IBGateway needs to be running on the local machine and listening on port 4002. Note: demo or paper account recommended!! 😏

using InteractiveBrokers

wrap = InteractiveBrokers.Wrapper(
         # Customized methods go here
         error= (err) ->
                  println("Error: $(something($(err.id), "NA")) $(err.errorCode) $(err.errorString) $(err.advancedOrderRejectJson)"),

         nextValidId= (orderId) -> println("Next OrderId: $orderId"),

         managedAccounts= (accountsList) -> println("Managed Accounts: $accountsList")

         # more method overrides can go here...
       );

# Connect to the server with clientId = 1
ib = InteractiveBrokers.connect(4001, 1);

# Start a background Task to process the server responses
InteractiveBrokers.start_reader(ib, wrap);

# Define contract
contract = InteractiveBrokers.Contract(symbol="GOOG",
                        secType="STK",
                        exchange="SMART",
                        currency="USD");

# Define order
order = InteractiveBrokers.Order();
order.action        = "BUY"
order.totalQuantity = 10
order.orderType     = "LMT"
order.lmtPrice      = 100

orderId = 1    # Should match whatever is returned by the server

# Send order
InteractiveBrokers.placeOrder(ib, orderId, contract, order)

# Disconnect
InteractiveBrokers.disconnect(ib)
Foreground vs. Background Processing

It is possible to process the server responses either within the main process or in a separate background Task:

  • foreground processing is triggered by invoking InteractiveBrokers.check_all(ib, wrap, Tab=Dict). It is the user's responsibility to call it on a regular basis, especially when data are streaming in.
  • background processing is started by InteractiveBrokers.start_reader(ib, wrap, Tab=Dict). A separate Task is started in the background, which monitors the connection and processes the responses as they arrive.

To avoid undesired effects, the two approaches should not be mixed together on the same connection.

Sink Format

Tab parameter of above examples is the sink format used when applicable. The library supports an extension for DataFrames (just pass DataFrame as a last parameter), otherwise Dict is the default format.

Implementation Details

The package does not export any name, therefore any functions or types described here need to be prefixed by InteractiveBrokers.*.

As Julia is not an object-oriented language, the functionality of the IB EClient class is provided here by regular functions. In particular:

  • connect(port, clientId, connectOptions): establish a connection and return a Connection object.
  • disconnect(::Connection): terminate the connection.
  • check_all(::Connection, ::Wrapper): process available responses, not blocking. Return the number of messages processed. Needs to be called regularly!
  • start_reader(::Connection, ::Wrapper): start a Task for background processing.
  • methods that send specific requests to the server. Refer to the official IB EClient class documentation for further details and method signatures. The only caveat is to remember to pass a Connection as first argument: e.g. reqContractDetails(ib::Connection, reqId:Int, contract::Contract)

Like the official IB EWrapper class, this struct holds the callbacks that are dispatched when responses are processed. The user provides the callback definitions as keyword arguments in the constructor, as shown above, and/or by setting the property of an existing instance.

A more comprehensive example is provided by simple_wrap(), which is used like this:

using InteractiveBrokers: InteractiveBrokers, Contract, reqContractDetails, simple_wrap, start_reader

data, wrap = simple_wrap();

ib = InteractiveBrokers.connect(4002, 1);
start_reader(ib, wrap);

reqContractDetails(ib, 99, Contract(conId=208813720, exchange="SMART"))

# Wait for the response and then access the "ContractDetails" result:
data[:cd]

Thanks to closures, data (a Dict in this case) is accessible by all wrap methods as well as the main program. This is one way to propagate incoming data to different parts of the program.

For more details about callback definitions and signatures, refer to the official IB EWrapper class documentation. As reference, the exact signatures used in this package are found here.

Notes

Callbacks are generally invoked with arguments and types matching the signatures as described in the official documentation. However, there are few exceptions:

  • tickPrice() has an extra size::Float64 argument, which is meaningful only when TickType ∈ {BID, ASK, LAST}. In these cases, the official IB API fires an extra tickSize() event instead.
  • historicalData() is invoked only once per request, presenting all the historical data as a single DataFrame, whereas the official IB API invokes it row-by-row.
  • scannerData() is also invoked once per request and its arguments are in fact vectors rather than single values.

These modifications make it possible to establish the rule: one callback per server response.

Consequently, historicalDataEnd() and scannerDataEnd() are redundant and are not used in this package.

DataFrame are passed to several other callbacks, such as: mktDepthExchanges(), smartComponents(), newsProviders(), histogramData(), marketRule() and the historicalTicks*() family.

Missing Values

Occasionally, for numerical types, there is the need to represent the lack of a value.

IB API does not have a uniform solution across the board, but rather it adopts a variety of sentinel values. They can be either the plain 0 or the largest representable value of a given type such as 2147483647 and 9223372036854775807 for 32- and 64-bit integers respectively or 1.7976931348623157E308 for 64-bit floating point.

This package makes an effort to use Julia built-in Nothing in all circumstances.

Data Structures

Other classes that mainly hold data are also replicated. They are implemented as Julia struct or mutable struct with names, types and default values matching the IB API counterparts: e.g. Contract, Order, ComboLeg, ExecutionFilter, ScannerSubscription and Condition*.

TagValueList are implemented as Julia NamedTuple. Wherever a TagValue is needed, something like this can be used:

tagvaluelist = (tag1="value1", tag2="value2")
# or, in case of an empty list:
emptylist = (;)

Values don't need to be of type String. Int and Float64 are also allowed.