Initial commit

Parses and executes substitute expressions (subexes)

So far subex has the following operations:
- Concatenation of a and b with ab
- Or with |
- Repeat maximally with *
- Repeat minimally with -
- Copy a specific character 'a'
- Copy any character '.'
- Store text matching a regex into slot 's': `$s(regex)`
- Output text in "" including loading from slots with '$'

Regexes support all the same operations as subexes minus storing and outputting

This first implementation gives very little thought to efficiency

Example:

./main 'according to all known laws of aviation' '$1(.-)$m(( .* )| ).*"$m$1"'

This swaps the first and last words of the input string

1 files changed, 123 insertions, 0 deletions

diff --git a/main/subexstate.go b/main/subexstate.go
new file mode 100644
index 0000000..cc697f0
--- /dev/null
+++ b/main/subexstate.go
@@ -0,0 +1,123 @@
+package main
+
+import (
+	"strings"
+)
+
+type SubexState interface {
+	eat(store Store, char rune) []SubexBranch
+	accepting(store Store) []string
+}
+
+type SubexGroupState struct {
+	first, second SubexState
+}
+func (state SubexGroupState) eat(store Store, char rune) []SubexBranch {
+	otherStore := store.clone()
+	return append(state.first.eat(store, char), state.second.eat(otherStore, char)...)
+}
+func (state SubexGroupState) accepting(store Store) []string {
+	return append(state.first.accepting(store), state.second.accepting(store)...)
+}
+
+type SubexStoreState struct {
+	match RegexState
+	slot rune
+	next SubexState
+	input string
+}
+func (state SubexStoreState) eat(store Store, char rune) []SubexBranch {
+	var nextStates []SubexBranch
+	if state.match.accepting() {
+		store[state.slot] = state.input
+		nextStates = state.next.eat(store, char)
+	}
+	nextRegexStates := state.match.eat(char)
+	for _, regexState := range nextRegexStates {
+		nextStates = append(nextStates, SubexBranch {
+			state: SubexStoreState {
+				match: regexState,
+				slot: state.slot,
+				next: state.next,
+				input: state.input + string(char),
+			},
+			output: "",
+			store: store.clone(),
+		})
+	}
+	return nextStates
+}
+func (state SubexStoreState) accepting(store Store) []string {
+	if state.match.accepting() {
+		return state.next.accepting(store)
+	}
+	return nil
+}
+
+type SubexOutputState struct {
+	content []TransducerOutput
+	next SubexState
+}
+func (state SubexOutputState) build(store Store) string {
+	var builder strings.Builder
+	for _, part := range state.content {
+		builder.WriteString(part.build(store))
+	}
+	return builder.String()
+}
+func (state SubexOutputState) eat(store Store, char rune) []SubexBranch {
+	content := state.build(store)
+	nextStates := state.next.eat(store, char)
+	for i := range nextStates {
+		nextStates[i].output = content + nextStates[i].output
+	}
+	return nextStates
+}
+func (state SubexOutputState) accepting(store Store) []string {
+	content := state.build(store)
+	outputs := state.next.accepting(store)
+	for i := range outputs {
+		outputs[i] = content + outputs[i]
+	}
+	return outputs
+}
+
+type SubexNoneState struct {}
+func (state SubexNoneState) eat(store Store, char rune) []SubexBranch {
+	return nil
+}
+func (state SubexNoneState) accepting(store Store) []string {
+	return []string{""}
+}
+
+type SubexCopyRuneState struct {
+	rune rune
+	next SubexState
+}
+func (state SubexCopyRuneState) eat(store Store, char rune) []SubexBranch {
+	if char == state.rune {
+		return []SubexBranch{{
+			state: state.next,
+			output: string(char),
+			store: store,
+		}}
+	}
+	return nil
+}
+func (state SubexCopyRuneState) accepting(store Store) []string {
+	return nil
+}
+
+type SubexCopyAnyState struct {
+	next SubexState
+}
+func (state SubexCopyAnyState) eat(store Store, char rune) []SubexBranch {
+	return []SubexBranch{{
+		state: state.next,
+		output: string(char),
+		store: store,
+	}}
+}
+func (state SubexCopyAnyState) accepting(store Store) []string {
+	return nil
+}