Allows you to run runtime text expressions on IObservable<T>.
Suppose a few scientists are working on a project of monitoring temperatures inside Amazon Rainforest, their temperature probe reports to satellite and all the temperature readings are received by a computer running the .NET Program that you are maintaining. You receive the temperature changes through IObservable (a Publish-Subscribe model)
IObservable<double> temperatureInFahrenheit = SomeMethodToGetTemperatureChanges();You are getting temperature in Fahrenheit, but scientists like Celcius better, you have a text field in your application that allows Scientists to write a formula f(t) so they can type experessions like the following.
( _ - 32 ) * 5/9
Underscore _ references the object yielded by input observable. In this case is the temperature values in double
You want to apply the above expression to an existing observable, CalcRx allows you to do the following.
IObservable<double> temperatureInCelcius = temperatureInFahrenheit.Evaluate<double, double>("( _ - 32 ) * 5/9");Now suppose you have a observable that gives you back the Stock Trades
IObservable<Trade> myTrades = SomeWebService.MyTrades();Traders that are using your application want to run a certain formula (like ones in Excel) on the Trades or Ticks that are coming in live. A formula such as SellPrice - BuyPrice
CalcRx can access properties of objects yielded by observables. So you could do
IObservable<double> profits = myTrades.Evaluate<Trade, double>("SellPrice - BuyPrice");Now, you want to have more fun and introduce functions
IObservable<int> numbers = Observable.Range(0, 10);
IObservable<double> squareRoots = numbers.Evaluate<int, double>("SQRT(_)",
new [] {new Function("SQRT", new Func<IObservable<int>, IObservable<double>>(
numberObservable => numberObservable.Select(Math.Sqrt)))});CalcRx has built-in Function Packs so you don't have to write some basic ones.
IObservable<int> series = Observable.Range(0, 10);
IObservable<double> runningSum = numbers.Evaluate<int, double>("SUM(_)", FunctionPacks.BasicAggregationFunctionPack);Similarly you can use a combination of function packs
IObservable<Trade> = NYSEService.GetTrades();
IObservable<double> last5MovingAvg = numbers.Evaluate<Trade, double>("SUM(REF(_, 5)) / 5",
FunctionPacks.BasicAggregationFunctionPack
.Concat(FunctionPacks.HistoryFunctionPack));The next few releases will address these issues
- Better syntax and runtime error handling and reporting
- Support for multi-line expressions
- Support for building function packs out of expression (thats gotta be easy)
The parser is written using Coco/R compiler generator. The ATG grammar file is included in the FormulaParser directory.
Currently the project is under development and I am adding features as I get time. If you would like to join in, feel free to fork it and send me the pull requests. If you need any support in compiling, or adding more features drop me a line.