Paper Artifacts

A Dataset of Simplified Syntax Trees for C# Sebastian Proksch, Sven Amann, Sarah Nadi, Mira Mezini

Links and Artifacts


You are allowed to download this dataset for scientific purposes only. The licenses of the contained source code are contained in the corresponding Github repositories.


How to use the dataset

You can read the IR from Java and C#, you will find the data bindings in our repository (Java, C#). These links point to the "Context", the root node of the IR. Please follow the types to find the rest of the model.

We provide helpers for the de-/serialization of this model in Java and in C#.

The dataset is easy to use. We provide an example file with several mini-snippets that explain how to read the dataset, traverse the IR with a visitor, access information, apply an inline transformation, or perform a points-to analysis. For details, please consult our examples for the correct usage of the CART platform.

If you want to create a separate dataset, you can do that, our tooling is published as open source. The source-code transformation is modelled as a ReSharper test case and has to be run from within Visual Studio. Because of licensing issues there is no way to run it externally, e.g., from the command line. To run your own analyses, you just have to adapt the path in our BulkAnalysisTest and start the test with the ReSharper test runner in Visual Studio. The test will store the generated intermediate representation (IR) in the location specified by you.