From bb39c49a0dbfd0ee05ec5ac3bba96916b175fa92 Mon Sep 17 00:00:00 2001
From: _ov4 <ov4@ov4.org>
Date: Tue, 8 Sep 2026 12:26:46 +0000
Subject: add guard system

---
 include/lexer/token.h | 58 +++++++++++++++++++++++++++++++++++++++++++--------
 module-test/ast.cc    | 15 +++++++++----
 2 files changed, 60 insertions(+), 13 deletions(-)

diff --git a/include/lexer/token.h b/include/lexer/token.h
index df7ef1b..a562ac6 100644
--- a/include/lexer/token.h
+++ b/include/lexer/token.h
@@ -102,7 +102,7 @@ bool token_continue(string s, char next)
 }
 
 // handle [prev, current)
-void token_push(string push_s, int bracket_depth)
+void token_push(string push_s, int bracket_depth, bool force_text = false)
 {
     T_token tmp;
     tmp.bracket_depth = bracket_depth;
@@ -111,13 +111,14 @@ void token_push(string push_s, int bracket_depth)
 
     println("finding {} in map", trimmed);
 
-    if (symbol_property.contains(trimmed)) // symbol
+    if (symbol_property.contains(trimmed) && !force_text) // symbol
     {
+        println("token_push: adding symbol");
         tmp.token_type = symbol_property[trimmed].enum_type;
-    
     } 
     else
     {
+        println("token_push: adding text");
         tmp.token_type = TEXT;
         tmp.text = trimmed;
     }
@@ -129,24 +130,63 @@ int tokenizer(string s)
     if (s.size() == 0) return -1;
 
     size_t len = s.size();
-    size_t prev = 0;
     size_t token_vector_index = 0;
     string prev_str = "";
-    bool is_in_symbol;
+    bool force_text = false;
     int bracket_depth = 0;
+    guard_type guard = GUARD_OFF;
 
 
     for (size_t i = 0; i < len; i++)
     {
+
+        char c = s[i];
+
+        prev_str += c;
+        
         println("---");
-        println("prev {} current char {}", prev, s[i]);
+        println("prev_str .{}. next char {}", prev_str, s[i+1]);
+
+        if (c == '\'' || c == '\"')
+        {
+            println("quotation");
+
+            // turn on GUARD
+            if (guard == GUARD_OFF)
+            {
+                guard = (c == '\'') ? GUARD_STRONG : GUARD_WEAK;
+                println("set guard {}", (c == '\'') ? "GUARD_STRONG" : "GUARD_WEAK");
+                continue;
+            }
+            // turn off
+            else if (guard == GUARD_STRONG && c == '\'')
+            {
+                guard = GUARD_OFF;
+                println("quitting strong guard");
+                force_text = true;
+            }
+            //turn off
+            else if (guard == GUARD_WEAK && c == '\"')
+            {
+                guard = GUARD_OFF;
+                println("quitting weak guard");
+                force_text = true;
+            } else
+                continue;
+        }
+        // not quotation mark
+        else if ((guard == GUARD_STRONG || guard == GUARD_WEAK))
+        {
+            println("tokenizer: guarded, continue");
+            continue;
+        }
 
-        prev_str += s[i];
-        bracket_depth_changer(s[i], bracket_depth);
+        bracket_depth_changer(c, bracket_depth);
 
         if ((i+1 == len) || !token_continue(prev_str, s[i+1]))
         {
-            token_push(prev_str, bracket_depth);
+            token_push(prev_str, bracket_depth, force_text);
+            force_text = false;
             prev_str = "";
             continue;
         }
diff --git a/module-test/ast.cc b/module-test/ast.cc
index 53f3b12..60d4b22 100644
--- a/module-test/ast.cc
+++ b/module-test/ast.cc
@@ -28,7 +28,7 @@ constexpr int MAXLINE = 1024;
 
 int main()
 {
-    int test_case = 7;
+    int test_case = 10;
     string s;
 
     switch (test_case)
@@ -63,13 +63,20 @@ int main()
             s = "((alpha&|&&beta)||gamma&&&delta)";
             break;
 
-        // guard test
+        // simple guard test
         case 8:
-            s = "\'foo\' && \"bar\"";
+            s = "\'text\' &&";
             break;
 
+        // guard test
         case 9:
-            s = "&(alpha)";
+            s = "\'realOperator\' && \"fake&&\" real ||";
+            break;
+
+        // guard test
+        // with space
+        case 10:
+            s = "\'This should be 1 token\' symbol && text \'&&\'";
             break;
 
         default:
-- 
cgit v1.3.1-4-g156e