feat: Add support for converting multiple NFAs into one DFA. #7
Conversation
| accept: Vec<State>, | ||
| states: HashSet<State>, | ||
| transitions: HashMap<State, HashMap<char, Transition>>, // from_state -> symbol -> to_state | ||
| dfa_to_accepted_nfa_state_mapping: Option<HashMap<State, Vec<(usize, crate::nfa::nfa::State)>>>, // to determine which NFA gets matched |
There was a problem hiding this comment.
I'm not sure why cargo fmt doesn't work, but this line exceeds the 100-char limit, right? Can we put the comment in a separate line before this line? (It's hard to navigate long lines without using mouse, lol)
|
|
||
| // Helper functions for converting multiple NFAs to a single DFA | ||
| impl DFA { | ||
| fn epsilon_closure( |
There was a problem hiding this comment.
Nit: Shouldn't this be a part of NFA?
| use std::process::id; | ||
|
|
||
| #[derive(Clone, Debug, Eq, Hash, PartialEq)] | ||
| struct State(String); |
There was a problem hiding this comment.
Why do we need the state to be a string?
| ); | ||
| } | ||
|
|
||
| fn simulate(&self, input: &str) -> (Option<HashSet<usize>>, bool) { |
There was a problem hiding this comment.
We need a different API that takes char by char since the lexer emits chars as tokens, we don't know the size of string in advance.
In this way, we might also need an API to reset the current simulation state.
src/dfa/dfa.rs
Outdated
| start: State, | ||
| accept: Vec<State>, | ||
| states: HashSet<State>, | ||
| transitions: HashMap<State, HashMap<char, Transition>>, // from_state -> symbol -> to_state |
There was a problem hiding this comment.
As discussed offline, there's a performance concern in this part; shall we add a TODO to keep track of the issue?
LinZhihao-723
changed the title
Description
Validation performed