nreco / lambdaparser

Runtime parser for string expressions (formulas, method calls). Builds dynamic LINQ expression tree and compiles it to lambda delegate.
http://www.nrecosite.com/
MIT License
307 stars 55 forks source link
dot-net dotnetcore evalexpr evaluator lambda-expressions lambda-interpreter math-expressions

NReco LambdaParser

Runtime parser for string expressions (formulas, method calls, properties/fields/arrays accessors). LambdaParser builds dynamic LINQ expression tree and compiles it to the lambda delegate. Types are resolved at run-time like in dynamic languages.

NuGet Windows x64 Linux
NuGet Release AppVeyor Tests

Nuget package: NReco.LambdaParser

var lambdaParser = new NReco.Linq.LambdaParser();

var varContext = new Dictionary<string,object>();
varContext["pi"] = 3.14M;
varContext["one"] = 1M;
varContext["two"] = 2M;
varContext["test"] = "test";
Console.WriteLine( lambdaParser.Eval("pi>one && 0<one ? (1+8)/3+1*two : 0", varContext) ); // --> 5
Console.WriteLine( lambdaParser.Eval("test.ToUpper()", varContext) ); // --> TEST

(see unit tests for more expression examples)

Custom values comparison

By default LambdaParser uses ValueComparer for values comparison. You can provide your own implementation or configure its option to get desired behaviour:

The UseCache property determines whether the LambdaParser should cache parsed expressions. By default, UseCache is set to true, meaning expressions are cached to improve performance for repeated evaluations of the same expression.

Therefore, using a singleton instance of LambdaParser is recommended, rather than creating a new instance each time.

You can disable caching by setting UseCache to false if you want to save memory, especially when evaluating a large number of unique expressions.

var lambdaParser = new LambdaParser();
lambdaParser.UseCache = false;

Who is using this?

NReco.LambdaParser is in production use at SeekTable.com and PivotData microservice (used for user-defined calculated cube members: formulas, custom formatting).

License

Copyright 2016-2024 Vitaliy Fedorchenko and contributors

Distributed under the MIT license