Add a comment consumer mechanism to MCAsmLexer

This allows clients to register an AsmCommentConsumer with the MCAsmLexer,
which receives a callback each time a comment is parsed.

Differential Revision: https://reviews.llvm.org/D27511



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@289036 91177308-0d34-0410-b5e6-96231b3b80d8

include/llvm/MC/MCParser/MCAsmLexer.h

@@ -128,6 +128,20 @@ class AsmToken {
   }
 };
 
+/// A callback class which is notified of each comment in an assembly file as
+/// it is lexed.
+class AsmCommentConsumer {
+public:
+  virtual ~AsmCommentConsumer() {};
+
+  /// Callback function for when a comment is lexed. Loc is the start of the
+  /// comment text (excluding the comment-start marker). CommentText is the text
+  /// of the comment, excluding the comment start and end markers, and the
+  /// newline for single-line comments.
+  virtual void HandleComment(SMLoc Loc, StringRef CommentText) = 0;
+};
+
+
 /// Generic assembler lexer interface, for use by target specific assembly
 /// lexers.
 class MCAsmLexer {
@@ -145,6 +159,7 @@ class MCAsmLexer {
   bool SkipSpace;
   bool AllowAtInIdentifier;
   bool IsAtStartOfStatement;
+  AsmCommentConsumer *CommentConsumer;
 
   MCAsmLexer();
 
@@ -234,6 +249,10 @@ class MCAsmLexer {
 
   bool getAllowAtInIdentifier() { return AllowAtInIdentifier; }
   void setAllowAtInIdentifier(bool v) { AllowAtInIdentifier = v; }
+
+  void setCommentConsumer(AsmCommentConsumer *CommentConsumer) {
+    this->CommentConsumer = CommentConsumer;
+  }
 };
 
 } // End llvm namespace

lib/MC/MCParser/AsmLexer.cpp

@@ -181,12 +181,19 @@ AsmToken AsmLexer::LexSlash() {
 
   // C Style comment.
   ++CurPtr;  // skip the star.
+  const char *CommentTextStart = CurPtr;
   while (CurPtr != CurBuf.end()) {
     switch (*CurPtr++) {
     case '*':
       // End of the comment?
       if (*CurPtr != '/')
         break;
+      // If we have a CommentConsumer, notify it about the comment.
+      if (CommentConsumer) {
+        CommentConsumer->HandleComment(
+            SMLoc::getFromPointer(CommentTextStart),
+            StringRef(CommentTextStart, CurPtr - 1 - CommentTextStart));
+      }
       ++CurPtr;   // End the */.
       return AsmToken(AsmToken::Comment,
                       StringRef(TokStart, CurPtr - TokStart));
@@ -202,10 +209,18 @@ AsmToken AsmLexer::LexLineComment() {
   // comment. While it would be nicer to leave this two tokens,
   // backwards compatability with TargetParsers makes keeping this in this form
   // better.
+  const char *CommentTextStart = CurPtr;
   int CurChar = getNextChar();
   while (CurChar != '\n' && CurChar != '\r' && CurChar != EOF)
     CurChar = getNextChar();
 
+  // If we have a CommentConsumer, notify it about the comment.
+  if (CommentConsumer) {
+    CommentConsumer->HandleComment(
+        SMLoc::getFromPointer(CommentTextStart),
+        StringRef(CommentTextStart, CurPtr - 1 - CommentTextStart));
+  }
+
   IsAtStartOfLine = true;
   // This is a whole line comment. leave newline
   if (IsAtStartOfStatement)

lib/MC/MCParser/MCAsmLexer.cpp

@@ -13,7 +13,8 @@
 using namespace llvm;
 
 MCAsmLexer::MCAsmLexer()
-    : TokStart(nullptr), SkipSpace(true), IsAtStartOfStatement(true) {
+    : TokStart(nullptr), SkipSpace(true), IsAtStartOfStatement(true),
+      CommentConsumer(nullptr) {
   CurTok.emplace_back(AsmToken::Space, StringRef());
 }