[Parser] Do not eagerly lex strings (#6543)

Lex them on demand instead to avoid wasted work.

2 files changed, 25 insertions, 49 deletions

diff --git a/src/parser/lexer.cpp b/src/parser/lexer.cpp
index 8384b5047..f3646c0be 100644
--- a/src/parser/lexer.cpp
+++ b/src/parser/lexer.cpp
@@ -1004,17 +1004,6 @@ std::optional<float> Token::getF32() const {
   return {};
 }
 
-std::optional<std::string_view> Token::getString() const {
-  if (auto* tok = std::get_if<StringTok>(&data)) {
-    if (tok->str) {
-      return std::string_view(*tok->str);
-    }
-    // Remove quotes.
-    return span.substr(1, span.size() - 2);
-  }
-  return {};
-}
-
 void Lexer::skipSpace() {
   while (true) {
     if (auto ctx = annotation(next())) {
@@ -1054,6 +1043,22 @@ bool Lexer::takeRParen() {
   return false;
 }
 
+std::optional<std::string> Lexer::takeString() {
+  if (curr) {
+    return std::nullopt;
+  }
+  if (auto result = str(next())) {
+    index += result->span.size();
+    advance();
+    if (result->str) {
+      return result->str;
+    }
+    // Remove quotes.
+    return std::string(result->span.substr(1, result->span.size() - 2));
+  }
+  return std::nullopt;
+}
+
 std::optional<Name> Lexer::takeID() {
   if (curr) {
     return std::nullopt;
@@ -1132,8 +1137,6 @@ void Lexer::lexToken() {
     tok = Token{t->span, IntTok{t->n, t->sign}};
   } else if (auto t = float_(next())) {
     tok = Token{t->span, FloatTok{t->nanPayload, t->d}};
-  } else if (auto t = str(next())) {
-    tok = Token{t->span, StringTok{t->str}};
   } else {
     // TODO: Do something about lexing errors.
     curr = std::nullopt;
@@ -1204,15 +1207,6 @@ std::ostream& operator<<(std::ostream& os, const FloatTok& tok) {
   return os << tok.d;
 }
 
-std::ostream& operator<<(std::ostream& os, const StringTok& tok) {
-  if (tok.str) {
-    os << '"' << *tok.str << '"';
-  } else {
-    os << "(raw string)";
-  }
-  return os;
-}
-
 std::ostream& operator<<(std::ostream& os, const Token& tok) {
   std::visit([&](const auto& t) { os << t; }, tok.data);
   return os << " \"" << tok.span << "\"";
diff --git a/src/parser/lexer.h b/src/parser/lexer.h
index 0fe0faa27..64d1fc6be 100644
--- a/src/parser/lexer.h
+++ b/src/parser/lexer.h
@@ -68,16 +68,8 @@ struct FloatTok {
   friend std::ostream& operator<<(std::ostream&, const FloatTok&);
 };
 
-struct StringTok {
-  // If the string contains escapes, this is its contents.
-  std::optional<std::string> str;
-
-  bool operator==(const StringTok& other) const { return str == other.str; }
-  friend std::ostream& operator<<(std::ostream&, const StringTok&);
-};
-
 struct Token {
-  using Data = std::variant<IntTok, FloatTok, StringTok>;
+  using Data = std::variant<IntTok, FloatTok>;
   std::string_view span;
   Data data;
 
@@ -90,7 +82,6 @@ struct Token {
   template<typename T> std::optional<T> getI() const;
   std::optional<double> getF64() const;
   std::optional<float> getF32() const;
-  std::optional<std::string_view> getString() const;
 
   bool operator==(const Token&) const;
   friend std::ostream& operator<<(std::ostream& os, const Token&);
@@ -145,6 +136,10 @@ public:
       if (peekLParen() || peekRParen()) {
         return true;
       }
+      // Do not count the parentheses in strings.
+      if (takeString()) {
+        continue;
+      }
       if (!curr) {
         ++index;
       }
@@ -218,27 +213,14 @@ public:
     return std::nullopt;
   }
 
-  std::optional<std::string> takeString() {
-    if (curr) {
-      if (auto s = curr->getString()) {
-        std::string ret(*s);
-        advance();
-        return ret;
-      }
-    }
-    return {};
-  }
+  std::optional<std::string> takeString();
 
   std::optional<Name> takeName() {
-    // TODO: Move this to lexer and validate UTF.
+    // TODO: Validate UTF.
     if (auto str = takeString()) {
-      // Copy to a std::string to make sure we have a null terminator, otherwise
-      // the `Name` constructor won't work correctly.
-      // TODO: Update `Name` to use string_view instead of char* and/or to take
-      // rvalue strings to avoid this extra copy.
-      return Name(std::string(*str));
+      return Name(*str);
     }
-    return {};
+    return std::nullopt;
   }
 
   bool takeSExprStart(std::string_view expected) {