Fix rdar://6562329, a static analyzer crash Ted noticed on

wine sources.  This was happening because HighlightMacros was
calling EnterMainFile multiple times on the same preprocessor
object and getting an assert due to the new #line stuff (the
file in question was bison output with #line directives).

The fix for this is to not reenter the file.  Instead,
relex the tokens in raw mode, swizzle them a bit and repreprocess
the token stream.  An added bonus of this is that rewrite macros
will now hilight the macro definition as well as its uses.  Woo.

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@64480 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Lex/Preprocessor.cpp b/lib/Lex/Preprocessor.cpp
index 69aa7cb51a6225f4631acc0801339be54fa56785..f16d83c5a2e2c46fe198f4d6f2c3b0737d214ac0 100644 (file)
--- a/lib/Lex/Preprocessor.cpp
+++ b/lib/Lex/Preprocessor.cpp
@@ -626,7 +626,10 @@ static void InitializePredefinedMacros(Preprocessor &PP,
 /// EnterMainSourceFile - Enter the specified FileID as the main source file,
 /// which implicitly adds the builtin defines etc.
 void Preprocessor::EnterMainSourceFile() {
-  
+  // We do not allow the preprocessor to reenter the main file.  Doing so will
+  // cause FileID's to accumulate information from both runs (e.g. #line
+  // information) and predefined macros aren't guaranteed to be set properly.
+  assert(NumEnteredSourceFiles == 0 && "Cannot reenter the main file!");
   FileID MainFileID = SourceMgr.getMainFileID();
   
   // Enter the main file source buffer.

diff --git a/lib/Rewrite/HTMLRewrite.cpp b/lib/Rewrite/HTMLRewrite.cpp
index c0f3937600752876352238b0f430a357da494b7f..a0f289b2b4fa3af7343c59c5175c159360668818 100644 (file)
--- a/lib/Rewrite/HTMLRewrite.cpp
+++ b/lib/Rewrite/HTMLRewrite.cpp
@@ -344,8 +344,8 @@ void html::AddHeaderFooterInternalBuiltinCSS(Rewriter& R, FileID FID,
 void html::SyntaxHighlight(Rewriter &R, FileID FID, Preprocessor &PP) {
   RewriteBuffer &RB = R.getEditBuffer(FID);
 
-  const SourceManager &SourceMgr = PP.getSourceManager();
-  Lexer L(FID, SourceMgr, PP.getLangOptions());
+  const SourceManager &SM = PP.getSourceManager();
+  Lexer L(FID, SM, PP.getLangOptions());
   const char *BufferStart = L.getBufferStart();
   
   // Inform the preprocessor that we want to retain comments as tokens, so we 
@@ -360,7 +360,7 @@ void html::SyntaxHighlight(Rewriter &R, FileID FID, Preprocessor &PP) {
   while (Tok.isNot(tok::eof)) {
     // Since we are lexing unexpanded tokens, all tokens are from the main
     // FileID.
-    unsigned TokOffs = SourceMgr.getFileOffset(Tok.getLocation());
+    unsigned TokOffs = SM.getFileOffset(Tok.getLocation());
     unsigned TokLen = Tok.getLength();
     switch (Tok.getKind()) {
     default: break;
@@ -398,7 +398,7 @@ void html::SyntaxHighlight(Rewriter &R, FileID FID, Preprocessor &PP) {
       unsigned TokEnd = TokOffs+TokLen;
       L.LexFromRawLexer(Tok);
       while (!Tok.isAtStartOfLine() && Tok.isNot(tok::eof)) {
-        TokEnd = SourceMgr.getFileOffset(Tok.getLocation())+Tok.getLength();
+        TokEnd = SM.getFileOffset(Tok.getLocation())+Tok.getLength();
         L.LexFromRawLexer(Tok);
       }
       
@@ -416,23 +416,55 @@ void html::SyntaxHighlight(Rewriter &R, FileID FID, Preprocessor &PP) {
 }
 
 /// HighlightMacros - This uses the macro table state from the end of the
-/// file, to reexpand macros and insert (into the HTML) information about the
+/// file, to re-expand macros and insert (into the HTML) information about the
 /// macro expansions.  This won't be perfectly perfect, but it will be
 /// reasonably close.
 void html::HighlightMacros(Rewriter &R, FileID FID, Preprocessor& PP) {
   
   RewriteBuffer &RB = R.getEditBuffer(FID);
   
+  // Re-lex the raw token stream into a token buffer.
+  const SourceManager &SM = PP.getSourceManager();
+  std::vector<Token> TokenStream;
+  
+  Lexer L(FID, SM, PP.getLangOptions());
+  
+  // Lex all the tokens in raw mode, to avoid entering #includes or expanding
+  // macros.
+  while (1) {
+    Token Tok;
+    L.LexFromRawLexer(Tok);
+    
+    // If this is a # at the start of a line, discard it from the token stream.
+    // We don't want the re-preprocess step to see #defines, #includes or other
+    // preprocessor directives.
+    if (Tok.is(tok::hash) && Tok.isAtStartOfLine())
+      continue;
+    
+    // If this raw token is an identifier, the raw lexer won't have looked up
+    // the corresponding identifier info for it.  Do this now so that it will be
+    // macro expanded when we re-preprocess it.
+    if (Tok.is(tok::identifier)) {
+      // Change the kind of this identifier to the appropriate token kind, e.g.
+      // turning "for" into a keyword.
+      Tok.setKind(PP.LookUpIdentifierInfo(Tok)->getTokenID());
+    }    
+      
+    TokenStream.push_back(Tok);
+    
+    if (Tok.is(tok::eof)) break;
+  }
+  
   // Inform the preprocessor that we don't want comments.
   PP.SetCommentRetentionState(false, false);
-  
-  // Start parsing the specified input file.
-  PP.EnterMainSourceFile();
+
+  // Enter the tokens we just lexed.  This will cause them to be macro expanded
+  // but won't enter sub-files (because we removed #'s).
+  PP.EnterTokenStream(&TokenStream[0], TokenStream.size(), false, false);
   
   TokenConcatenation ConcatInfo(PP);
   
   // Lex all the tokens.
-  const SourceManager &SourceMgr = PP.getSourceManager();
   Token Tok;
   PP.Lex(Tok);
   while (Tok.isNot(tok::eof)) {
@@ -443,8 +475,8 @@ void html::HighlightMacros(Rewriter &R, FileID FID, Preprocessor& PP) {
     }
     
     // Ignore tokens whose instantiation location was not the main file.
-    SourceLocation LLoc = SourceMgr.getInstantiationLoc(Tok.getLocation());
-    std::pair<FileID, unsigned> LLocInfo = SourceMgr.getDecomposedLoc(LLoc);
+    SourceLocation LLoc = SM.getInstantiationLoc(Tok.getLocation());
+    std::pair<FileID, unsigned> LLocInfo = SM.getDecomposedLoc(LLoc);
     
     if (LLocInfo.first != FID) {
       PP.Lex(Tok);
@@ -457,7 +489,7 @@ void html::HighlightMacros(Rewriter &R, FileID FID, Preprocessor& PP) {
     // Get the size of current macro call itself.
     // FIXME: This should highlight the args of a function-like
     // macro, using a heuristic.
-    unsigned TokLen = Lexer::MeasureTokenLength(LLoc, SourceMgr);
+    unsigned TokLen = Lexer::MeasureTokenLength(LLoc, SM);
     
     unsigned TokOffs = LLocInfo.second;
     // Highlight the macro invocation itself.
@@ -476,7 +508,7 @@ void html::HighlightMacros(Rewriter &R, FileID FID, Preprocessor& PP) {
     // instantiation.  It would be really nice to pop up a window with all the
     // spelling of the tokens or something.
     while (!Tok.is(tok::eof) &&
-           SourceMgr.getInstantiationLoc(Tok.getLocation()) == LLoc) {
+           SM.getInstantiationLoc(Tok.getLocation()) == LLoc) {
       // Insert a newline if the macro expansion is getting large.
       if (LineLen > 60) {
         Expansion += "<br>";

diff --git a/test/Misc/emit-html.c b/test/Misc/emit-html.c
new file mode 100644 (file)
index 0000000..1bcab12
--- /dev/null
+++ b/test/Misc/emit-html.c
@@ -0,0 +1,5 @@
+// RUN: clang %s -emit-html -o -
+
+// rdar://6562329
+#line 42 "foo.c"
+