Added fields to state, to determine capturing group information. 0th group refers to entire match

master
Aadhavan Srinivasan 2 weeks ago
parent 822d1f319f
commit 11f7f1d746

@ -33,6 +33,10 @@ type State struct {
except []rune // Only valid if allChars is true - match all characters _except_ the ones in this block. Useful for inverting character classes. 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 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 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
groupNum int // Which capturing group the node starts / ends
} }
// Clones the NFA starting from the given state. // Clones the NFA starting from the given state.
@ -65,6 +69,9 @@ func cloneStateHelper(state *State, cloneMap map[*State]*State) *State {
allChars: state.allChars, allChars: state.allChars,
except: append([]rune{}, state.except...), except: append([]rune{}, state.except...),
lookaroundRegex: state.lookaroundRegex, lookaroundRegex: state.lookaroundRegex,
groupEnd: state.groupEnd,
groupBegin: state.groupBegin,
groupNum: state.groupNum,
} }
cloneMap[state] = clone cloneMap[state] = clone
for i, s := range state.output { for i, s := range state.output {
@ -113,17 +120,17 @@ func (s State) checkAssertion(str []rune, idx int) bool {
// 2. Run it on the test string. // 2. Run it on the test string.
// 3. Based on the kind of lookaround (and the indices we get), determine what action to take. // 3. Based on the kind of lookaround (and the indices we get), determine what action to take.
startState := s.lookaroundNFA startState := s.lookaroundNFA
matchIndices := findAllMatches(startState, str) matchIndices := findAllMatches(startState, str, startState.lookaroundNumCaptureGroups)
numMatchesFound := 0 numMatchesFound := 0
for _, matchIdx := range matchIndices { for _, matchIdx := range matchIndices {
if s.assert == PLA || s.assert == NLA { // Lookahead - return true (or false) if at least one match starts at the current index if s.assert == PLA || s.assert == NLA { // Lookahead - return true (or false) if at least one match starts at the current index
if matchIdx.startIdx == idx { if matchIdx[0].startIdx == idx {
numMatchesFound++ numMatchesFound++
} }
} }
if s.assert == PLB || s.assert == NLB { // Lookbehind - return true (or false) if at least one match _ends_ at the current index. if s.assert == PLB || s.assert == NLB { // Lookbehind - return true (or false) if at least one match _ends_ at the current index.
if matchIdx.endIdx == idx { if matchIdx[0].endIdx == idx {
numMatchesFound++ numMatchesFound++
} }
} }

Loading…
Cancel
Save