lambda_calculus 1.3.2

A library for lambda calculus. It supports parsing and evaluating lambda terms with different strategies. It also has preliminary support for type inference.

1.3.1

• Fix a concurrency bug in type inference.

• The current type inference implementation is very nasty and inefficient. I will rewrite it in the future.

1.3.0

• Modify the parsing syntax:

  1. Stop allowing abstractions without delimiters, e.g. λx x is not allowed anymore, use λx. x or λx -> x instead.
  2. Allow spaces between the lambda symbol and the variable, e.g. λ x -> x is now allowed. It was already allowed according to the previous documentation, but the implementation was not consistent.
  3. Allow inner abstractions to be omitted, e.g. λx. λy. x y can be written as λx y. x y.
  4. Stop allowing special depth variables such as _x1 or _y2. These underscore variables are not prohibited.

• Improve the pretty-printing of Lambda to not creating variables with underscores. Instead it will choose appropriate fresh names that are guaranteed to not conflict with existing variables.

• Fix bugs in the parser.

• Make LambdaBuilder and Lambda final.

• More documentation.

1.2.0

• Change the signature of fmap so that "non-leaf" callbacks no longer have access to the term itself. This is because previous usage of this term is not consistent.

• Fix the problem where variables with the same name can conflict with each other during evaluation.

• Better printing format for LambdaType.

• Fix typos in documentation.

• Make Lambda final.

• Remove dependency on dartz.

• More tests.

1.1.1

• Better documentation.

1.1.0

• Stop allowing constructing Lambda directly, instead, use LambdaBuilder.

• Access constants via LambdaBuilder.constants or Lambda.constants.

• Fix the problem where the index of a variable is sometimes null.

1.0.0

• Initial version.