diff --git a/regex/nfa.go b/regex/nfa.go index 994dc96..839af20 100644 --- a/regex/nfa.go +++ b/regex/nfa.go @@ -22,18 +22,18 @@ const ( alwaysTrueAssert // An assertion that is always true ) -type State struct { +type state struct { content stateContents // Contents of current state isEmpty bool // If it is empty - Union operator and Kleene star states will be empty isLast bool // If it is the last state (acept state) - output []*State // The outputs of the current state ie. the 'outward arrows'. A union operator state will have more than one of these. - transitions map[int][]*State // Transitions to different states (maps a character (int representation) to a _list of states. This is useful if one character can lead multiple states eg. ab|aa) + output []*state // The outputs of the current state ie. the 'outward arrows'. A union operator state will have more than one of these. + transitions map[int][]*state // Transitions to different states (maps a character (int representation) to a _list of states. This is useful if one character can lead multiple states eg. ab|aa) isKleene bool // Identifies whether current node is a 0-state representing Kleene star assert assertType // Type of assertion of current node - NONE means that the node doesn't assert anything allChars bool // Whether or not the state represents all characters (eg. a 'dot' metacharacter). A 'dot' node doesn't store any contents directly, as it would take up too much space except []rune // Only valid if allChars is true - match all characters _except_ the ones in this block. Useful for inverting character classes. lookaroundRegex string // Only for lookaround states - Contents of the regex that the lookaround state holds - lookaroundNFA *State // Holds the NFA of the lookaroundRegex - if it exists + lookaroundNFA *state // Holds the NFA of the lookaroundRegex - if it exists lookaroundNumCaptureGroups int // Number of capturing groups in lookaround regex if current node is a lookaround groupBegin bool // Whether or not the node starts a capturing group groupEnd bool // Whether or not the node ends a capturing group @@ -44,14 +44,14 @@ type State struct { } // Clones the NFA starting from the given state. -func cloneState(start *State) *State { - return cloneStateHelper(start, make(map[*State]*State)) +func cloneState(start *state) *state { + return cloneStateHelper(start, make(map[*state]*state)) } // Helper function for clone. The map is used to keep track of which states have // already been copied, and which ones haven't. // This function was created using output from Llama3.1:405B. -func cloneStateHelper(stateToClone *State, cloneMap map[*State]*State) *State { +func cloneStateHelper(stateToClone *state, cloneMap map[*state]*state) *state { // Base case - if the clone exists in our map, return it. if clone, exists := cloneMap[stateToClone]; exists { return clone @@ -61,12 +61,12 @@ func cloneStateHelper(stateToClone *State, cloneMap map[*State]*State) *State { } // Recursive case - if the clone doesn't exist, create it, add it to the map, // and recursively call for each of the transition states. - clone := &State{ + clone := &state{ content: append([]int{}, stateToClone.content...), isEmpty: stateToClone.isEmpty, isLast: stateToClone.isLast, - output: make([]*State, len(stateToClone.output)), - transitions: make(map[int][]*State), + output: make([]*state, len(stateToClone.output)), + transitions: make(map[int][]*state), isKleene: stateToClone.isKleene, assert: stateToClone.assert, zeroMatchFound: stateToClone.zeroMatchFound, @@ -86,7 +86,7 @@ func cloneStateHelper(stateToClone *State, cloneMap map[*State]*State) *State { } } for k, v := range stateToClone.transitions { - clone.transitions[k] = make([]*State, len(v)) + clone.transitions[k] = make([]*state, len(v)) for i, s := range v { if s == stateToClone { clone.transitions[k][i] = clone @@ -104,7 +104,7 @@ func cloneStateHelper(stateToClone *State, cloneMap map[*State]*State) *State { // Checks if the given state's assertion is true. Returns true if the given // state doesn't have an assertion. -func (s State) checkAssertion(str []rune, idx int) bool { +func (s state) checkAssertion(str []rune, idx int) bool { if s.assert == alwaysTrueAssert { return true } @@ -171,7 +171,7 @@ func (s State) checkAssertion(str []rune, idx int) bool { } // Returns true if the contents of 's' contain the value at the given index of the given string -func (s State) contentContains(str []rune, idx int) bool { +func (s state) contentContains(str []rune, idx int) bool { if s.assert != noneAssert { return s.checkAssertion(str, idx) } @@ -182,19 +182,19 @@ func (s State) contentContains(str []rune, idx int) bool { return slices.Contains(s.content, int(str[idx])) } -func (s State) isLookaround() bool { +func (s state) isLookaround() bool { return s.assert == plaAssert || s.assert == plbAssert || s.assert == nlaAssert || s.assert == nlbAssert } // Returns the matches for the character at the given index of the given string. // Also returns the number of matches. Returns -1 if an assertion failed. -func (s State) matchesFor(str []rune, idx int) ([]*State, int) { +func (s state) matchesFor(str []rune, idx int) ([]*state, int) { // Assertions can be viewed as 'checks'. If the check fails, we return // an empty array and 0. // If it passes, we treat it like any other state, and return all the transitions. if s.assert != noneAssert { if s.checkAssertion(str, idx) == false { - return make([]*State, 0), -1 + return make([]*state, 0), -1 } } listTransitions := s.transitions[int(str[idx])] @@ -211,39 +211,39 @@ func (s State) matchesFor(str []rune, idx int) ([]*State, int) { } // verifyLastStatesHelper performs the depth-first recursion needed for verifyLastStates -func verifyLastStatesHelper(state *State, visited map[*State]bool) { - if len(state.transitions) == 0 { - state.isLast = true +func verifyLastStatesHelper(st *state, visited map[*state]bool) { + if len(st.transitions) == 0 { + st.isLast = true return } // if len(state.transitions) == 1 && len(state.transitions[state.content]) == 1 && state.transitions[state.content][0] == state { // Eg. a* - if len(state.transitions) == 1 { // Eg. a* + if len(st.transitions) == 1 { // Eg. a* var moreThanOneTrans bool // Dummy variable, check if all the transitions for the current's state's contents have a length of one - for _, c := range state.content { - if len(state.transitions[c]) != 1 || state.transitions[c][0] != state { + for _, c := range st.content { + if len(st.transitions[c]) != 1 || st.transitions[c][0] != st { moreThanOneTrans = true } } - state.isLast = !moreThanOneTrans + st.isLast = !moreThanOneTrans } - if state.isKleene { // A State representing a Kleene Star has transitions going out, which loop back to it. If all those transitions point to the same (single) state, then it must be a last state - transitionDests := make([]*State, 0) - for _, v := range state.transitions { + if st.isKleene { // A State representing a Kleene Star has transitions going out, which loop back to it. If all those transitions point to the same (single) state, then it must be a last state + transitionDests := make([]*state, 0) + for _, v := range st.transitions { transitionDests = append(transitionDests, v...) } if allEqual(transitionDests...) { - state.isLast = true + st.isLast = true return } } - if visited[state] == true { + if visited[st] == true { return } - visited[state] = true - for _, states := range state.transitions { + visited[st] = true + for _, states := range st.transitions { for i := range states { - if states[i] != state { + if states[i] != st { verifyLastStatesHelper(states[i], visited) } } @@ -251,12 +251,12 @@ func verifyLastStatesHelper(state *State, visited map[*State]bool) { } // verifyLastStates enables the 'isLast' flag for the leaf nodes (last states) -func verifyLastStates(start []*State) { - verifyLastStatesHelper(start[0], make(map[*State]bool)) +func verifyLastStates(start []*state) { + verifyLastStatesHelper(start[0], make(map[*state]bool)) } // Concatenates s1 and s2, returns the start of the concatenation. -func concatenate(s1 *State, s2 *State) *State { +func concatenate(s1 *state, s2 *state) *state { if s1 == nil { return s2 } @@ -269,13 +269,13 @@ func concatenate(s1 *State, s2 *State) *State { return s1 } -func kleene(s1 State) (*State, error) { +func kleene(s1 state) (*state, error) { if s1.isEmpty && s1.assert != noneAssert { return nil, fmt.Errorf("previous token is not quantifiable") } - toReturn := &State{} - toReturn.transitions = make(map[int][]*State) + toReturn := &state{} + toReturn.transitions = make(map[int][]*state) toReturn.content = newContents(EPSILON) toReturn.isEmpty = true toReturn.isKleene = true @@ -291,9 +291,9 @@ func kleene(s1 State) (*State, error) { return toReturn, nil } -func alternate(s1 *State, s2 *State) *State { - toReturn := &State{} - toReturn.transitions = make(map[int][]*State) +func alternate(s1 *state, s2 *state) *state { + toReturn := &state{} + toReturn.transitions = make(map[int][]*state) toReturn.output = append(toReturn.output, s1.output...) toReturn.output = append(toReturn.output, s2.output...) // Unique append is used here (and elsewhere) to ensure that, @@ -313,9 +313,9 @@ func alternate(s1 *State, s2 *State) *State { return toReturn } -func question(s1 *State) *State { // Use the fact that ab? == a(b|) - s2 := &State{} - s2.transitions = make(map[int][]*State) +func question(s1 *state) *state { // Use the fact that ab? == a(b|) + s2 := &state{} + s2.transitions = make(map[int][]*state) s2.content = newContents(EPSILON) s2.output = append(s2.output, s2) s2.isEmpty = true @@ -324,10 +324,10 @@ func question(s1 *State) *State { // Use the fact that ab? == a(b|) } // Creates and returns a new state with the 'default' values. -func newState() State { - ret := State{ - output: make([]*State, 0), - transitions: make(map[int][]*State), +func newState() state { + ret := state{ + output: make([]*state, 0), + transitions: make(map[int][]*state), assert: noneAssert, except: append([]rune{}, 0), lookaroundRegex: "", @@ -339,7 +339,7 @@ func newState() State { } // Creates and returns a state that _always_ has a zero-length match. -func zeroLengthMatchState() State { +func zeroLengthMatchState() state { start := newState() start.content = newContents(EPSILON) start.isEmpty = true