From 365183e243d164185bca6ad9fa4e0d75664800f4 Mon Sep 17 00:00:00 2001
From: Roland Reichwein <mail@reichwein.it>
Date: Sat, 21 Mar 2020 22:11:57 +0100
Subject: C++ grammer (WIP)

---
 Makefile     |  2 +-
 cpp.cpp      | 86 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++----
 grammer.cpp  |  2 ++
 test-cpp.cpp |  9 +++++--
 4 files changed, 91 insertions(+), 8 deletions(-)

diff --git a/Makefile b/Makefile
index b8f5495..b28f356 100644
--- a/Makefile
+++ b/Makefile
@@ -3,7 +3,7 @@ PROJECTNAME=minicc
 CXX=clang++-9
 #CXX=g++-8
 
-CXXFLAGS=-O0 -D_DEBUG
+CXXFLAGS=-O0 -g -D_DEBUG
 # -fprofile-instr-generate -fcoverage-mapping
 # gcc:--coverage
 #CXXFLAGS=-O2 -DNDEBUG
diff --git a/cpp.cpp b/cpp.cpp
index d877564..69ca847 100644
--- a/cpp.cpp
+++ b/cpp.cpp
@@ -99,7 +99,85 @@ namespace {
  std::unordered_set<std::string> keywords {
   "alignas",
   "alignof",
-  // ... Keywords table, p.15
+  "asm",
+  "auto",
+  "bool",
+  "break",
+  "case",
+  "catch",
+  "char",
+  "char8_t",
+  "char16_t",
+  "char32_t",
+  "class",
+  "concept",
+  "const",
+  "consteval",
+  "constexpr",
+  "constinit",
+  "const_cast",
+  "continue",
+  "co_await",
+  "co_return",
+  "co_yield",
+  "decltype",
+  "default",
+  "delete",
+  "do",
+  "double",
+  "dynamic_cast",
+  "else",
+  "enum",
+  "explicit",
+  "export",
+  "extern",
+  "false",
+  "float",
+  "for",
+  "friend",
+  "goto",
+  "if",
+  "inline",
+  "int",
+  "long",
+  "mutable",
+  "namespace",
+  "new",
+  "noexcept",
+  "nullptr",
+  "operator",
+  "private",
+  "protected",
+  "public",
+  "register",
+  "reinterpret_cast",
+  "requires",
+  "return",
+  "short",
+  "signed",
+  "sizeof",
+  "static",
+  "static_assert",
+  "static_cast",
+  "struct",
+  "switch",
+  "template",
+  "this",
+  "thread_local",
+  "throw",
+  "true",
+  "try",
+  "typedef",
+  "typeid",
+  "typename",
+  "union",
+  "unsigned",
+  "using",
+  "virtual",
+  "void",
+  "volatile",
+  "wchar_t",
+  "while",
  };
 }
 
@@ -119,12 +197,10 @@ std::vector<Token> CPP::tokens_from_pptokens(std::vector<Token> pp_tokens)
  for (auto& token: pp_tokens) {
   if (pp_types.find(token.type) != pp_types.end()) {
    if (token.type == "identifier") {
-#if 0
     if (keywords.find(token.value) != keywords.end())
-     result.emplace_back("keyword", token.value);
+     result.emplace_back(token.value, token.value);
     else
-#endif
-    result.emplace_back(Token{"identifier"s, token.value});
+     result.emplace_back(Token{"identifier"s, token.value});
    }
    else if (token.type == "preprocessing-op-or-punc")
     result.emplace_back(Token{token.value, token.value});
diff --git a/grammer.cpp b/grammer.cpp
index 2379e9a..5557b12 100644
--- a/grammer.cpp
+++ b/grammer.cpp
@@ -230,6 +230,8 @@ bool Compiler::AddRootNode()
    const auto& variants{bnf[type]};
    for (int i = 0; i < variants.size(); i++) {
     const std::vector<std::string> & variant{variants[i]};
+    if (variant.size() == 0)
+     continue; // TODO: Handle case of empty rule (see e.g. C++ "attribute-list")
     if (child_type == variant[0]) {
      if (node_type == "") {
       node_type = type;
diff --git a/test-cpp.cpp b/test-cpp.cpp
index d08f9b0..2a67b38 100644
--- a/test-cpp.cpp
+++ b/test-cpp.cpp
@@ -24,7 +24,7 @@ class CppTest: public ::testing::Test
 {
 protected:
  CppTest() {
-  //debug = true;
+  debug = true;
  }
  ~CppTest() {
  }
@@ -40,8 +40,13 @@ TEST_F(CppTest, preprocessing_tokenize) {
  auto tokens = cpp.tokens_from_pptokens(pp_tokens);
 
  ASSERT_EQ(tokens.size(), 9);
+#if 0
+ for (auto &i: tokens) {
+  std::cout << i.type << ": " << i.value << std::endl;
+ }
+#endif
 
- //auto result = cpp.analysis(tokens);
+ auto result = cpp.analysis(tokens);
 }
 #endif
 
-- 
cgit v1.2.3