From c6fe8cccc49a70af8b8367fcfe19ff8a05f16d7a Mon Sep 17 00:00:00 2001
From: Roland Reichwein <mail@reichwein.it>
Date: Sun, 19 Jan 2020 19:58:33 +0100
Subject: Added Locations

---
 TODO       |  1 -
 minicc.cpp | 93 +++++++++++++++++++++++++++++++++++++++++++++++---------------
 2 files changed, 71 insertions(+), 23 deletions(-)

diff --git a/TODO b/TODO
index 751699a..e69de29 100644
--- a/TODO
+++ b/TODO
@@ -1 +0,0 @@
-Locations in source: File:line
diff --git a/minicc.cpp b/minicc.cpp
index 37fc974..f548181 100644
--- a/minicc.cpp
+++ b/minicc.cpp
@@ -4,6 +4,7 @@
 #include "gtest/gtest.h"
 
 #include <algorithm>
+#include <cctype>
 #include <deque>
 #include <map>
 #include <memory>
@@ -69,6 +70,13 @@ public:
   node_num = 0;
  }
 
+ // Type of lexical token
+ std::string GetType() {
+  if (node_num > 0 && nodes[root].child_names.size() == 1)
+   return nodes[root].child_names[0];
+  return "";
+ }
+
  bool Valid(const std::string& Top) const {
   // A token is non empty
   if (node_num == 0)
@@ -295,6 +303,31 @@ public:
 
 };
 
+struct Location {
+ size_t line;
+ size_t column;
+};
+
+bool operator==(const Location &a, const Location &b)
+{
+ return (a.line == b.line && a.column == b.column);
+}
+
+struct Token {
+ std::string type;
+ std::string value;
+ Location location;
+};
+
+bool operator==(const Token &a, const Token &b)
+{
+ return (a.type == b.type && a.value == b.value && a.location == b.location);
+}
+
+std::ostream& operator<<(std::ostream& os, const Token& token) {
+ return os << token.type << ": " << token.value << "(" << token.location.line << ":" << token.location.column << ")";
+}
+
 class Lexer
 {
 
@@ -302,14 +335,16 @@ private:
  const BNF &bnf;
  const std::string& Top;
 
+ Location location{1, 0};
+
  std::map<std::string, std::set<std::string>> ReverseBNF;
 
  // to be called on token end
- void FinalizeTree(Tree& tree, std::string& token, std::vector<std::string>& result)
+ void FinalizeTree(Tree& tree, std::string& token, std::vector<Token>& result)
  {
   tree.Resolve(bnf, ReverseBNF);
   if (tree.Valid(Top)) {
-   result.push_back(token);
+   result.emplace_back(Token{tree.GetType(), token, Location{location.line, location.column - token.size()}});
    token.clear();
   }
   tree.clear();
@@ -320,9 +355,9 @@ public:
  {
  }
 
- std::vector<std::string> Lex(const std::string& s)
+ std::vector<Token> Lex(const std::string& s)
  {
-  std::vector<std::string> result;
+  std::vector<Token> result;
   std::string token;
 
   std::string Whitespace{"\t \n\r"};
@@ -330,7 +365,14 @@ public:
 
   for (size_t pos{0}; pos < s.size(); pos++) {
    char c{s[pos]};
-   std::cout << "Char: |" << c << "|" << std::endl;
+   if (c == '\n') {
+    location.column = 0;
+    location.line++;
+   } else if (std::isprint(c)) {
+    location.column++;
+   }
+
+   //std::cout << "Char: |" << c << "|" << std::endl;
    if (Whitespace.find(c) != std::string::npos) { // found whitespace character
     // evaluate token up to now and skip whitespace
     FinalizeTree(tree, token, result);
@@ -353,23 +395,11 @@ public:
 
 };
 
-ProgramNode Compile(std::vector<std::string> Tokens, std::string Top, BNF bnf, Terminals terminals)
+ProgramNode Compile(std::vector<Token> Tokens, std::string Top, BNF bnf, Terminals terminals)
 {
- BNF ReverseBNF;//{ Reverse(bnf)};
+ std::map<std::string, std::set<std::string>> ReverseBNF{ Reverse(bnf)};
 
  if (Tokens.size()){
-  std::string Token = Tokens[0];
-#if 0
-  auto Path = GetPath(Token, ReverseBNF, Top, terminals);
-  if (Path.size()) {
-   size_t Index{1};
-   while (Index < Tokens.size()) {
-    Path = GetPath(Token, ReverseBNF, Top, terminals, Path);
-    Index++;
-   }
-  } else
-   throw std::runtime_error("Invalid token: "s + Token);
-#endif
  } else
   throw std::runtime_error("No tokens!");
 
@@ -416,15 +446,34 @@ TEST_F(Test, BNF) {
  std::set<std::string> Terminals{"identifier", "=", ";"};
 
  std::string Code{"a = bc ; c = 123 ; esd = Ff ; 1 = XYZ"};
+ std::vector<Token> tokens_reference{
+  {"identifier", "a", { 1, 1} },
+  {"preprocessing-op-or-punc", "=", { 1, 3}},
+  {"identifier", "bc", { 1, 5}},
+  {"preprocessing-op-or-punc", ";", { 1, 8}},
+  {"identifier", "c", { 1, 10}},
+  {"preprocessing-op-or-punc", "=", { 1, 12}},
+  {"pp-number", "123", { 1, 14}},
+  {"preprocessing-op-or-punc", ";", { 1, 18}},
+  {"identifier", "esd", { 1, 20}},
+  {"preprocessing-op-or-punc", "=", { 1, 24}},
+  {"identifier", "Ff", { 1, 26}},
+  {"preprocessing-op-or-punc", ";", { 1, 29}},
+  {"pp-number", "1", { 1, 31}},
+  {"preprocessing-op-or-punc", "=", { 1, 33}},
+  {"identifier", "XYZ", { 1, 34}},
+ };
 
  Lexer lexer(LexBNF, LexTop);
  auto tokens = lexer.Lex(Code);
-#if 1
+
+ ASSERT_EQ(tokens, tokens_reference);
+#if 0
  for (const auto& i: tokens) {
-  std::cout << i << std::endl;
+  std::cout << i.value << std::endl;
  }
 #endif
- //auto Program = Compile(tokens, Top, bnf, Terminals);
+ auto Program = Compile(tokens, Top, bnf, Terminals);
 }
 
 int main(int argc, char* argv[]) {
-- 
cgit v1.2.3