From 190941c3043297b00fc125a1de7e0ff7497f6f0f Mon Sep 17 00:00:00 2001
From: Christian Parpart <christian@parpart.family>
Date: Tue, 11 Apr 2023 10:07:43 +0200
Subject: [PATCH] wip

Signed-off-by: Christian Parpart <christian@parpart.family>
---
 src/regex_dfa/Lexable.h      | 10 ++++++----
 src/regex_dfa/Lexer-inl.h    | 21 +--------------------
 src/regex_dfa/Lexer_test.cpp |  5 +++--
 3 files changed, 10 insertions(+), 26 deletions(-)

diff --git a/src/regex_dfa/Lexable.h b/src/regex_dfa/Lexable.h
index 3384edb7b6..343408d3f7 100644
--- a/src/regex_dfa/Lexable.h
+++ b/src/regex_dfa/Lexable.h
@@ -322,7 +322,7 @@ inline Token LexerIterator<Token, Machine, RequiresBeginOfLine, Trace>::recogniz
     stack.push_back(BadState);
 
     if constexpr (Trace)
-        tracef("recognize: startState {}, offset {} {}",
+        tracef("recognizeOne: startState {}, offset {} {}",
                stateName(state),
                offset_,
                isBeginOfLine_ ? "BOL" : "no-BOL");
@@ -331,6 +331,7 @@ inline Token LexerIterator<Token, Machine, RequiresBeginOfLine, Trace>::recogniz
     while (state != ErrorState)
     {
         Symbol ch = nextChar(); // one of: input character, ERROR or EOF
+        fmt::print("recognizeOne: ch: {}\n", ch);
         currentToken_.literal.push_back(ch);
 
         // we do not stack.clear() stack if isAcceptState(state) as we need this information iff
@@ -344,7 +345,7 @@ inline Token LexerIterator<Token, Machine, RequiresBeginOfLine, Trace>::recogniz
     while (state != BadState && !isAcceptState(state))
     {
         if constexpr (Trace)
-            tracef("recognize: backtrack: current state {} {}; stack: {}",
+            tracef("recognizeOne: backtrack: current state {} {}; stack: {}",
                    stateName(state),
                    isAcceptState(state) ? "accepting" : "non-accepting",
                    toString(stack));
@@ -391,7 +392,7 @@ inline Token LexerIterator<Token, Machine, RequiresBeginOfLine, Trace>::recogniz
                currentToken_.offset,
                offset_,
                quotedString(currentToken_.literal),
-               quoted(currentChar_));
+               prettySymbol(currentChar_));
 
     if (!isAcceptState(state))
         throw LexerError { offset_ };
@@ -464,7 +465,7 @@ inline Symbol LexerIterator<Token, Machine, RequiresBeginOfLine, Trace>::nextCha
     }
 
     int ch = source_->get();
-    fmt::print("source.get: => {} (0x{:02X}, {})\n", ch, (uint8_t)ch, prettySymbol(ch));
+    fmt::print("source.get: => {} (0x{:02X}, {})\n", ch, (uint16_t)ch, prettySymbol(ch));
     if (ch < 0)
     {
         currentChar_ = Symbols::EndOfFile;
@@ -490,6 +491,7 @@ inline void LexerIterator<Token, Machine, RequiresBeginOfLine, Trace>::rollback(
     {
         offset_--;
         buffered_.push_back(currentToken_.literal.back());
+        tracef("Lexer:{}: rollback '{}'", offset_, prettySymbol(buffered_.back()));
     }
 }
 
diff --git a/src/regex_dfa/Lexer-inl.h b/src/regex_dfa/Lexer-inl.h
index fc897daec9..b925548a05 100644
--- a/src/regex_dfa/Lexer-inl.h
+++ b/src/regex_dfa/Lexer-inl.h
@@ -17,25 +17,6 @@
 namespace regex_dfa
 {
 
-static inline std::string quoted(int ch)
-{
-    if (ch == Symbols::Epsilon)
-        return "ε";
-    if (ch == Symbols::Error)
-        return "Error";
-    if (ch == Symbols::BeginOfLine)
-        return "BOL";
-    if (ch == Symbols::EndOfLine)
-        return "EOL";
-    if (ch == Symbols::EndOfFile)
-        return "EOF";
-    if (ch == '\n')
-        return "\\n";
-    if (ch == ' ')
-        return "\\s";
-    return fmt::format("{}", ch);
-}
-
 static inline std::string quotedString(const std::string& s)
 {
     std::stringstream sstr;
@@ -255,7 +236,7 @@ inline Token Lexer<Token, Machine, RequiresBeginOfLine, Debug>::recognizeOne()
                oldOffset_,
                offset_,
                quotedString(word_),
-               quoted(currentChar_));
+               prettySymbol(currentChar_));
 
     if (!isAcceptState(state))
         throw LexerError { offset_ };
diff --git a/src/regex_dfa/Lexer_test.cpp b/src/regex_dfa/Lexer_test.cpp
index 2b2df35482..e0e3a4634e 100644
--- a/src/regex_dfa/Lexer_test.cpp
+++ b/src/regex_dfa/Lexer_test.cpp
@@ -114,7 +114,7 @@ TEST_CASE("regex_Lexable.one")
     auto src = Lexable<LookaheadToken, StateId, false, true> { ld,
                                                                make_unique<stringstream>("abba abcdef"),
                                                                [](const string& msg) {
-                                                                   UNSCOPED_INFO(msg);
+                                                                   fmt::print("trace: {}\n", msg);
                                                                } };
     auto lexer = begin(src);
     auto eof = end(src);
@@ -168,8 +168,9 @@ TEST_CASE("regex_Lexer.match_eol")
     cc.parse(RULES);
 
     LexerDef ld = cc.compile();
+    INFO(fmt::format("LexerDef:\n{}", ld.to_string()));
     Lexable<LookaheadToken, StateId, false, true> ls { ld, "abba eol\nabba", [](const string& msg) {
-                                                          INFO(msg);
+                                                          fmt::print("trace: {}\n", msg);
                                                       } };
     auto lexer = begin(ls);