Fix an infinite loop in comment lexer: we were not advancing in the input character...

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@159303 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/AST/CommentLexer.cpp b/lib/AST/CommentLexer.cpp
index 0b76050ff0bc92edfc6df6ca125d17e5a78bfe8d..c3a801d924c24b944f129b9c7b7752f11a6a9881 100644 (file)
--- a/lib/AST/CommentLexer.cpp
+++ b/lib/AST/CommentLexer.cpp
@@ -357,6 +357,11 @@ void Lexer::lexCommentText(Token &T) {
           setupAndLexHTMLOpenTag(T);
         else if (C == '/')
           lexHTMLCloseTag(T);
+        else {
+          StringRef Text(BufferPtr, TokenPtr - BufferPtr);
+          formTokenWithChars(T, TokenPtr, tok::text);
+          T.setText(Text);
+        }
         return;
       }
 

diff --git a/unittests/AST/CommentLexer.cpp b/unittests/AST/CommentLexer.cpp
index a1e2427880cf9533b48cbfaa5f7691200e0b18e2..5b4712df2599760a9d8f4ad16a60c7b4a3a2115d 100644 (file)
--- a/unittests/AST/CommentLexer.cpp
+++ b/unittests/AST/CommentLexer.cpp
@@ -683,6 +683,28 @@ TEST_F(CommentLexerTest, HTML1) {
 }
 
 TEST_F(CommentLexerTest, HTML2) {
+  const char *Source =
+    "// < tag";
+
+  std::vector<Token> Toks;
+
+  lexString(Source, Toks);
+
+  ASSERT_EQ(4U, Toks.size());
+
+  ASSERT_EQ(tok::text,         Toks[0].getKind());
+  ASSERT_EQ(StringRef(" "),    Toks[0].getText());
+
+  ASSERT_EQ(tok::text,         Toks[1].getKind());
+  ASSERT_EQ(StringRef("<"),    Toks[1].getText());
+
+  ASSERT_EQ(tok::text,         Toks[2].getKind());
+  ASSERT_EQ(StringRef(" tag"), Toks[2].getText());
+
+  ASSERT_EQ(tok::newline,      Toks[3].getKind());
+}
+
+TEST_F(CommentLexerTest, HTML3) {
   const char *Sources[] = {
     "// <tag",
     "// <tag "
@@ -705,7 +727,7 @@ TEST_F(CommentLexerTest, HTML2) {
   }
 }
 
-TEST_F(CommentLexerTest, HTML3) {
+TEST_F(CommentLexerTest, HTML4) {
   const char *Source = "// <tag=";
 
   std::vector<Token> Toks;
@@ -726,7 +748,7 @@ TEST_F(CommentLexerTest, HTML3) {
   ASSERT_EQ(tok::newline,       Toks[3].getKind());
 }
 
-TEST_F(CommentLexerTest, HTML4) {
+TEST_F(CommentLexerTest, HTML5) {
   const char *Sources[] = {
     "// <tag attr",
     "// <tag attr "
@@ -752,7 +774,7 @@ TEST_F(CommentLexerTest, HTML4) {
   }
 }
 
-TEST_F(CommentLexerTest, HTML5) {
+TEST_F(CommentLexerTest, HTML6) {
   const char *Sources[] = {
     "// <tag attr=",
     "// <tag attr ="
@@ -780,7 +802,7 @@ TEST_F(CommentLexerTest, HTML5) {
   }
 }
 
-TEST_F(CommentLexerTest, HTML6) {
+TEST_F(CommentLexerTest, HTML7) {
   const char *Sources[] = {
     "// <tag attr=\"",
     "// <tag attr = \"",
@@ -813,7 +835,7 @@ TEST_F(CommentLexerTest, HTML6) {
   }
 }
 
-TEST_F(CommentLexerTest, HTML7) {
+TEST_F(CommentLexerTest, HTML8) {
   const char *Source = "// <tag attr=@";
 
   std::vector<Token> Toks;
@@ -839,7 +861,7 @@ TEST_F(CommentLexerTest, HTML7) {
   ASSERT_EQ(tok::newline,       Toks[5].getKind());
 }
 
-TEST_F(CommentLexerTest, HTML8) {
+TEST_F(CommentLexerTest, HTML9) {
   const char *Sources[] = {
     "// <tag attr=\"val\\\"\\'val",
     "// <tag attr=\"val\\\"\\'val\"",
@@ -872,7 +894,7 @@ TEST_F(CommentLexerTest, HTML8) {
   }
 }
 
-TEST_F(CommentLexerTest, HTML9) {
+TEST_F(CommentLexerTest, HTML10) {
   const char *Sources[] = {
     "// <tag attr=\"val\\\"\\'val\">",
     "// <tag attr=\'val\\\"\\'val\'>"
@@ -905,7 +927,7 @@ TEST_F(CommentLexerTest, HTML9) {
   }
 }
 
-TEST_F(CommentLexerTest, HTML10) {
+TEST_F(CommentLexerTest, HTML11) {
   const char *Source = "// </";
 
   std::vector<Token> Toks;
@@ -924,7 +946,7 @@ TEST_F(CommentLexerTest, HTML10) {
 }
 
 
-TEST_F(CommentLexerTest, HTML11) {
+TEST_F(CommentLexerTest, HTML12) {
   const char *Source = "// </@";
 
   std::vector<Token> Toks;
@@ -945,7 +967,7 @@ TEST_F(CommentLexerTest, HTML11) {
   ASSERT_EQ(tok::newline,        Toks[3].getKind());
 }
 
-TEST_F(CommentLexerTest, HTML12) {
+TEST_F(CommentLexerTest, HTML13) {
   const char *Sources[] = {
     "// </tag",
     "// </tag>",