A language that will hopefully be accepted as the project for the TKOM course. Clearly unfinished.
Important: This project will be developed on a hidden repo on WUT gitlab for now because of grading requirements. Whenever I have time I will try to update this repo with info from that other one, but they require a more rigid repo structure so I can't just copy this one. I could probably do it nicely with some git magic, but cant be bothered.
In the proj.py file there are two functions (for now):
lexeme_generator:TextIOStream, list -> Generator[Token]:- Consumes the text stream and returns a generator of tokens. Any lexing errors will be appended to the list.
parse_tokens:Generator[Token], list -> AST:- Consumes the generator and returns the AST without resolving names. All
unresolved names are
NameNodeinstances. Parsing errors are appended to the list. Unfinished.
There is also a temporary helper method test that takes a string and tries
to parse it, returning an error list if any errors are encountered.
The AST dataclass is probably what you're going to be most interested in at
this point of the project. It holds four dictionaries: one for each of words,
types, type aliases and type constructors. It also holds a list of all
definitions and top level comments in the sequential field, still debating
whether this is useful in any way.
For now it doesn't execute anything yet, but if you input this implementation of
fibonacci into test it should spit out the AST for it:
: fib ( n -- f(n) )
0 1 rot match:
case: Succ dip: over + swap ; ;
case: Zero drop ; ;