Choreia

A static analyzer to generate Choreography Automata from Go source.

What is

Aim of this project is to write a complete static analyzer that will parse and extract metadata from a Go source file and from that extrapolate a Choreography Automata, that is a Finite State Automta (FSA) that represents how the different Goroutines interact with each other during the program execution flow.

Some of the use cases of such program could be:

• Debugging pouposes: The program will output a SVG file that represents Choreography Automata and this can be a visual aid to debugging complex distributed systems
• Correctness and well-formedness: If the final Choreography Automata fullfills some properties such as Correctness and Well-formedness that in turn allows for some assumption to be made about our program and the interactions occuring between its parts
• Visual interpretation of protocols: This tool can be used to visualize both standard and proprietary comunication-based protocol, to do so we need only a mock implementation of our protocol in Go and the tool will take care of the visualization

How to

Before running the project you need to install some dependencies with the go get command, after that you can run the project with:

usr@computer:~/Choreia$ go run choreia/main.go -i input_file.go

or optionally build it and then running it as a standalone executable:

usr@computer:~/Choreia$ go build o your_path choreia/main.go
usr@computer:~/Choreia$ ./your_path choreia/main.go -i input_file.go

the latter is especially indicated when parsing large files as the compilation increases the execution speed. In addition -i arg to indicate the input file, other CLI argument are the following:

Shorthand	Extended	Usage	Default
-i	--input	The path to the Go source file
-o	--output	The path to where the data wll be saved	./choreia_out
-t	--trace	Prints to the stdout a trace of the AST while parsing
-h	--help	Show help message and usage instructions

Credits & Licensing

This project was made by me as Bachelor's degree Thesis for the Computer Science course at University of Bologna. Special thanks to the professor Ivan Lanese and to each contributors of the library used as dependencies.