add string literals.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@73858 91177308-0d34-0410-b5e6-96231b3b80d8

tools/llvm-mc/AsmLexer.cpp

@@ -77,7 +77,7 @@ asmtok::TokKind AsmLexer::LexIdentifier() {
   while (isalnum(*CurPtr) || *CurPtr == '_' || *CurPtr == '$' ||
          *CurPtr == '.' || *CurPtr == '@')
     ++CurPtr;
-  CurStrVal.assign(TokStart, CurPtr);   // Skip %
+  CurStrVal.assign(TokStart, CurPtr);   // Include %
   return asmtok::Identifier;
 }
 
@@ -194,6 +194,28 @@ asmtok::TokKind AsmLexer::LexDigit() {
   return asmtok::IntVal;
 }
 
+/// LexQuote: String: "..."
+asmtok::TokKind AsmLexer::LexQuote() {
+  int CurChar = getNextChar();
+  // TODO: does gas allow multiline string constants?
+  while (CurChar != '"') {
+    if (CurChar == '\\') {
+      // Allow \", etc.
+      CurChar = getNextChar();
+    }
+    
+    if (CurChar == EOF) {
+      PrintError(TokStart, "unterminated string constant");
+      return asmtok::Eof;
+    }
+
+    CurChar = getNextChar();
+  }
+  
+  CurStrVal.assign(TokStart, CurPtr);   // include quotes.
+  return asmtok::String;
+}
+
 
 asmtok::TokKind AsmLexer::LexToken() {
   TokStart = CurPtr;
@@ -228,6 +250,7 @@ asmtok::TokKind AsmLexer::LexToken() {
   case '%': return LexPercent();
   case '/': return LexSlash();
   case '#': return LexHash();
+  case '"': return LexQuote();
   case '0': case '1': case '2': case '3': case '4':
   case '5': case '6': case '7': case '8': case '9':
     return LexDigit();

tools/llvm-mc/AsmLexer.h

@@ -28,10 +28,15 @@ namespace asmtok {
     // Markers
     Eof, Error,
 
+    // String values.
     Identifier,
     Register,
+    String,
+    
+    // Integer values.
     IntVal,
     
+    // No-value.
     EndOfStatement,
     Colon,
     Plus,
@@ -70,7 +75,8 @@ public:
   asmtok::TokKind getKind() const { return CurKind; }
   
   const std::string &getCurStrVal() const {
-    assert((CurKind == asmtok::Identifier || CurKind == asmtok::Register) &&
+    assert((CurKind == asmtok::Identifier || CurKind == asmtok::Register ||
+            CurKind == asmtok::String) &&
            "This token doesn't have a string value");
     return CurStrVal;
   }
@@ -95,6 +101,7 @@ private:
   asmtok::TokKind LexSlash();
   asmtok::TokKind LexHash();
   asmtok::TokKind LexDigit();
+  asmtok::TokKind LexQuote();
 };
   
 } // end namespace llvm

tools/llvm-mc/llvm-mc.cpp

@@ -82,6 +82,9 @@ static int AssembleInput(const char *ProgName) {
     case asmtok::Register:
       outs() << "register: " << Lexer.getCurStrVal() << '\n';
       break;
+    case asmtok::String:
+      outs() << "string: " << Lexer.getCurStrVal() << '\n';
+      break;
     case asmtok::IntVal:
       outs() << "int: " << Lexer.getCurIntVal() << '\n';
       break;