From 6a6462e48072a24671dc0065b0e1c20a0b839096 Mon Sep 17 00:00:00 2001
From: Yufan Song <songyufan2003@163.com>
Date: Fri, 11 Oct 2024 16:07:44 +0800
Subject: [PATCH 1/2] Extend the semantics of parentheses

---
 nfa/construction.py | 73 ++++++++++++++++++++++++++-------------------
 1 file changed, 43 insertions(+), 30 deletions(-)

diff --git a/nfa/construction.py b/nfa/construction.py
index ce0cb13..7183f46 100644
--- a/nfa/construction.py
+++ b/nfa/construction.py
@@ -21,18 +21,17 @@ def pattern(pattern_string):
     lexer = Lexer(pattern_string)
     lexer.advance()
     nfa_pair = NfaPair()
-    group(nfa_pair)
+    expr(nfa_pair)
     # log_nfa(nfa_pair.start_node)
 
     return nfa_pair.start_node
 
 
 """
-group ::= ("(" expr ")")*
 expr ::= factor_conn ("|" factor_conn)*
 factor_conn ::= factor | factor factor*
 factor ::= (term | term ("*" | "+" | "?"))*
-term ::= char | "[" char "-" char "]" | .
+term ::= char | "[" char "-" char "]" | . | "(" expr ")"
 """
 
 
@@ -44,6 +43,8 @@ def term(pair_out):
         nfa_dot_char(pair_out)
     elif lexer.match(Token.CCL_START):
         nfa_set_nega_char(pair_out)
+    elif lexer.match(Token.OPEN_PAREN):
+        nfa_paren_around(pair_out)
 
 
 # 匹配单个字符
@@ -147,7 +148,6 @@ def factor_conn(pair_out):
 
 def is_conn(token):
     nc = [
-        Token.OPEN_PAREN,
         Token.CLOSE_PAREN,
         Token.AT_EOL,
         Token.EOS,
@@ -226,6 +226,19 @@ def nfa_option_closure(pair_out):
     return True
 
 
+# ()
+def nfa_paren_around(pair_out):
+    if not lexer.match(Token.OPEN_PAREN):
+        return False
+    
+    lexer.advance()
+    expr(pair_out)
+    if not lexer.match(Token.CLOSE_PAREN):
+        return False
+    lexer.advance()
+    return True
+
+
 def expr(pair_out):
     factor_conn(pair_out)
     pair = NfaPair()
@@ -246,32 +259,32 @@ def expr(pair_out):
     return True
 
 
-def group(pair_out):
-    if lexer.match(Token.OPEN_PAREN):
-        lexer.advance()
-        expr(pair_out)
-        if lexer.match(Token.CLOSE_PAREN):
-            lexer.advance()
-    elif lexer.match(Token.EOS):
-        return False
-    else: 
-        expr(pair_out)
-
-    while True:
-        pair = NfaPair()
-        if lexer.match(Token.OPEN_PAREN):
-            lexer.advance()
-            expr(pair)
-            pair_out.end_node.next_1 = pair.start_node
-            pair_out.end_node = pair.end_node
-            if lexer.match(Token.CLOSE_PAREN):
-                lexer.advance()
-        elif lexer.match(Token.EOS):
-            return False
-        else: 
-            expr(pair)
-            pair_out.end_node.next_1 = pair.start_node
-            pair_out.end_node = pair.end_node
+# def group(pair_out):
+#     if lexer.match(Token.OPEN_PAREN):
+#         lexer.advance()
+#         expr(pair_out)
+#         if lexer.match(Token.CLOSE_PAREN):
+#             lexer.advance()
+#     elif lexer.match(Token.EOS):
+#         return False
+#     else: 
+#         expr(pair_out)
+
+#     while True:
+#         pair = NfaPair()
+#         if lexer.match(Token.OPEN_PAREN):
+#             lexer.advance()
+#             expr(pair)
+#             pair_out.end_node.next_1 = pair.start_node
+#             pair_out.end_node = pair.end_node
+#             if lexer.match(Token.CLOSE_PAREN):
+#                 lexer.advance()
+#         elif lexer.match(Token.EOS):
+#             return False
+#         else: 
+#             expr(pair)
+#             pair_out.end_node.next_1 = pair.start_node
+#             pair_out.end_node = pair.end_node
 
     
     
\ No newline at end of file

From e9782173e31b003816ce40c9164c732766256e0d Mon Sep 17 00:00:00 2001
From: Yufan Song <songyufan2003@163.com>
Date: Fri, 11 Oct 2024 16:11:12 +0800
Subject: [PATCH 2/2] Add test cases for nested parentheses

---
 test/test.py | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/test/test.py b/test/test.py
index 23b7d86..a065a7a 100644
--- a/test/test.py
+++ b/test/test.py
@@ -16,6 +16,13 @@ def __init__(self, str, pattern, result):
 testLists.append(RegexMaterial("abbbbb", "[^c]+", True))
 testLists.append(RegexMaterial("ccccc", "[^c]+", False))
 testLists.append(RegexMaterial("123", "[1-3]+", True))
+testLists.append(RegexMaterial("ad", "a(bc)*d", True))
+testLists.append(RegexMaterial("abcd", "a(bc)*d", True))
+testLists.append(RegexMaterial("abcbcd", "a(bc)*d", True))
+testLists.append(RegexMaterial("abcdef", "a(b(cd)*e)?f", True))
+testLists.append(RegexMaterial("abef", "a(b(cd)*e)?f", True))
+testLists.append(RegexMaterial("af", "a(b(cd)*e)?f", True))
+testLists.append(RegexMaterial("abf", "a(b(cd)*e)?f", False))
 
 class TestRegex(unittest.TestCase):
     def test(self):