From 25cb79f01bb4b832eaff2a32bd0ddc9076f0b9f3 Mon Sep 17 00:00:00 2001
From: Aadhavan Srinivasan <aadhavan@twomorecents.org>
Date: Tue, 21 Jan 2025 22:10:07 -0500
Subject: [PATCH] Changed the value of EPSILON, so that we can use the NUL
 character (which it used to be) in a regex; Also added code to detect escaped
 backslashes

Specifically, I replace an escaped backslash with a metacharacter, then
replace it back later on. This prevents problems, like detecting whether
the opening bracket is escaped in '\\[a]'.
---
 compile.go | 12 ++++++++++++
 nfa.go     |  2 +-
 2 files changed, 13 insertions(+), 1 deletion(-)

diff --git a/compile.go b/compile.go
index c74cad0..7f43273 100644
--- a/compile.go
+++ b/compile.go
@@ -81,6 +81,10 @@ func shuntingYard(re string, flags ...ReFlag) ([]postfixNode, error) {
 	//
 	// Also check for non-capturing groups. The LPAREN of a non-capturing group looks like this: '(?:'
 	// I take this out, and put in a special character - NONCAPLPAREN_CHAR.
+	//
+	// Finally, check for escaped backslashes. Replace these with the BACKSLASH metacharacter. Later, in thompson(),
+	// these will be converted back. This avoids confusiuon in detecting whether a character is escaped eg. detecting
+	// whether '\\[a]' has an escaped opening bracket (it doesn't).
 	for i := 0; i < len(re_runes_orig); i++ {
 		c := re_runes_orig[i]
 		if c == '<' && (i == 0 || (re_runes_orig[i-1] != '\\' && re_runes_orig[i-1] != '?')) {
@@ -115,6 +119,9 @@ func shuntingYard(re string, flags ...ReFlag) ([]postfixNode, error) {
 		} else if c == '(' && i < len(re_runes_orig)-2 && re_runes_orig[i+1] == '?' && re_runes_orig[i+2] == ':' {
 			re_runes = append(re_runes, NONCAPLPAREN_CHAR)
 			i += 2
+		} else if c == '\\' && i < len(re_runes_orig)-1 && re_runes_orig[i+1] == '\\' { // Escaped backslash
+			re_runes = append(re_runes, ESC_BACKSLASH)
+			i++
 		} else {
 			re_runes = append(re_runes, c)
 		}
@@ -671,6 +678,11 @@ func thompson(re []postfixNode) (Reg, error) {
 
 				}
 			}
+
+			// Replace ESC_BACKSLASH with actual backslash, so that we can actually check if we encounter it
+			replaceByValue([]int(state.content), int(ESC_BACKSLASH), '\\')
+			replaceByValue(state.except, ESC_BACKSLASH, '\\')
+
 			nfa = append(nfa, &state)
 		}
 		if c.nodetype == LPAREN || c.nodetype == RPAREN {
diff --git a/nfa.go b/nfa.go
index 9bc9ae9..5d1a4a4 100644
--- a/nfa.go
+++ b/nfa.go
@@ -4,7 +4,7 @@ import (
 	"slices"
 )
 
-const EPSILON int = 0
+const EPSILON int = 0xF0000
 
 type assertType int