From cf8e75fae7702855b3d81ed11b2fb480c31a7fde Mon Sep 17 00:00:00 2001
From: Julius Ikkala <julius.ikkala@gmail.com>
Date: Tue, 6 May 2025 20:56:18 +0300
Subject: Parse char literals as integers (#6989)

* Parse char literals as integers

* Fix formatting

* Parse escaped chars correctly

---------

Co-authored-by: Yong He <yonghe@outlook.com>
---
 source/compiler-core/slang-lexer.cpp | 38 +++++++++++++++++++++++++++++++++---
 1 file changed, 35 insertions(+), 3 deletions(-)

(limited to 'source/compiler-core/slang-lexer.cpp')

diff --git a/source/compiler-core/slang-lexer.cpp b/source/compiler-core/slang-lexer.cpp
index d19ea388d..358a0ceff 100644
--- a/source/compiler-core/slang-lexer.cpp
+++ b/source/compiler-core/slang-lexer.cpp
@@ -6,6 +6,7 @@
 //
 
 #include "core/slang-char-encode.h"
+#include "core/slang-string-escape-util.h"
 #include "slang-core-diagnostics.h"
 #include "slang-name.h"
 #include "slang-source-loc.h"
@@ -829,17 +830,48 @@ FloatingPointLiteralValue getFloatingPointLiteralValue(
     return value;
 }
 
-static void _lexStringLiteralBody(Lexer* lexer, char quote)
+IntegerLiteralValue getCharLiteralValue(Token const& token)
 {
+    String unquotedContent = StringEscapeUtil::unquote('\'', token.getContent());
+    StringBuilder unescaped(4);
+    auto escapeHandler = StringEscapeUtil::getHandler(StringEscapeUtil::Style::Cpp);
+    escapeHandler->appendUnescaped(unquotedContent.getUnownedSlice(), unescaped);
+
+    char const* cursor = unescaped.getBuffer();
+
+    IntegerLiteralValue codepoint = getUnicodePointFromUTF8([&]() { return *cursor++; });
+    return codepoint;
+}
+
+static void _lexStringLiteralBody(Lexer* lexer, char quote, bool singleChar)
+{
+    int len = 0;
     for (;;)
     {
         int c = _peek(lexer);
         if (c == quote)
         {
+            if (singleChar && len == 0)
+            { // Empty char literal - size must be exactly 1.
+                if (auto sink = lexer->getDiagnosticSink())
+                {
+                    sink->diagnose(_getSourceLoc(lexer), LexerDiagnostics::illegalCharacterLiteral);
+                }
+            }
             _advance(lexer);
             return;
         }
 
+        len++;
+
+        if (singleChar && len == 2)
+        { // Char literal about to have more than 1 char.
+            if (auto sink = lexer->getDiagnosticSink())
+            {
+                sink->diagnose(_getSourceLoc(lexer), LexerDiagnostics::illegalCharacterLiteral);
+            }
+        }
+
         switch (c)
         {
         case kEOF:
@@ -1346,12 +1378,12 @@ static TokenType _lexTokenImpl(Lexer* lexer)
 
     case '\"':
         _advance(lexer);
-        _lexStringLiteralBody(lexer, '\"');
+        _lexStringLiteralBody(lexer, '\"', false);
         return TokenType::StringLiteral;
 
     case '\'':
         _advance(lexer);
-        _lexStringLiteralBody(lexer, '\'');
+        _lexStringLiteralBody(lexer, '\'', true);
         return TokenType::CharLiteral;
 
 
-- 
cgit v1.2.3