Table of Contents

• 1. What is it?
• 2. Help
• 3. How to run?
  • 1. Using the shell script (build.sh)
  • 2. Manually.
• 4. What it can do?
• 5. Sample output

1. What is it?

This is a lambda calculus type checker and evaluator. It is an F# implementation of the algorithms and concepts presented by Emmanuel Chailloux and Romain Demangeon in the TAS (Typage et Analyse Statique) course at Sorbonne University.

2. Help

ml-core 1.0.0
Copyright (C) 2022 ml-core

  -f, --file           File path to the file to evaluate and infer. If no file is specified a mini-REPL will run
                       instead.

  -l, --include-lib    (Default: false) If enabeled, a library file containing some predefined functions (map, range
                       ...) will be loaded.

  -d, --debug          (Default: false) Displays alpha-conversion and detailed evaluations.

  --help               Display this help screen.

  --version            Display version information.

3. How to run?

1. Manually (Recommanded)

1. Install the dotnet SDK and runtime here.

2. Go to the root folder (where the readme file is located).

3. Restore the projects :

   dotnet restore

4. To run the REPL :

   • Go to the ml-core folder :

     cd ml-core

   • run the project :

     dotnet run -- [your flags] (refer to the help section above)

   • example :

     dotnet run -- --include-lib --debug

     

5. To run the tests :

   • Go to the ml-tests folder :

     cd ml-tests

   • run the tests :

     dotnet test

     

6. To run the web app :

   • Go to the ml-web folder :

     cd ml-web

   • run the project :

     dotnet run

     A web browser should open. If not check the console to find the url 😄

     

     Example :

     

2. Using the shell script (build.sh)

1. Install the dotnet SDK and runtime here.

2. Go to the root folder (where the readme file is located).

3. Run the shell script using the source command (IMPORTANT).

   source build.sh

   This will generate the binary under the out folder.

4. WITHOUT changing directory, run the REPL :

   ml-core [your flags] (refer to the help section above)

   OR use a file

   ml-core -f [your file path] [your flags] (refer to the help section above)

   there is already a program.zfs file defnied in the root folder, you can update it and run it using the following command :

   ml-core -f program.zfs [your flags] (refer to the help section above)

   sample :

   

4. What it can do?

Refer to the rapport.pdf file for more details !

5. Sample output

Here are some sample outputs with the REPL :