Eat the UTF-8 BOM at the beginning of a file since it's ignored anyhow.

Nom Nom Nom.

Patch by Anton Korobeynikov!


git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@129174 91177308-0d34-0410-b5e6-96231b3b80d8

2 files changed, 19 insertions, 7 deletions

diff --git a/lib/Basic/SourceManager.cpp b/lib/Basic/SourceManager.cpp
index b6939ec7d5..8262feba7f 100644
--- a/lib/Basic/SourceManager.cpp
+++ b/lib/Basic/SourceManager.cpp
@@ -126,13 +126,12 @@ const llvm::MemoryBuffer *ContentCache::getBuffer(Diagnostic &Diag,
     if (Invalid) *Invalid = true;
     return Buffer.getPointer();
   }
-  
+
   // If the buffer is valid, check to see if it has a UTF Byte Order Mark
-  // (BOM).  We only support UTF-8 without a BOM right now.  See
+  // (BOM).  We only support UTF-8 with and without a BOM right now.  See
   // http://en.wikipedia.org/wiki/Byte_order_mark for more information.
   llvm::StringRef BufStr = Buffer.getPointer()->getBuffer();
-  const char *BOM = llvm::StringSwitch<const char *>(BufStr)
-    .StartsWith("\xEF\xBB\xBF", "UTF-8")
+  const char *InvalidBOM = llvm::StringSwitch<const char *>(BufStr)
     .StartsWith("\xFE\xFF", "UTF-16 (BE)")
     .StartsWith("\xFF\xFE", "UTF-16 (LE)")
     .StartsWith("\x00\x00\xFE\xFF", "UTF-32 (BE)")
@@ -145,9 +144,9 @@ const llvm::MemoryBuffer *ContentCache::getBuffer(Diagnostic &Diag,
     .StartsWith("\x84\x31\x95\x33", "GB-18030")
     .Default(0);
 
-  if (BOM) {
+  if (InvalidBOM) {
     Diag.Report(Loc, diag::err_unsupported_bom)
-      << BOM << ContentsEntry->getName();
+      << InvalidBOM << ContentsEntry->getName();
     Buffer.setInt(Buffer.getInt() | InvalidFlag);
   }
   
diff --git a/lib/Lex/Lexer.cpp b/lib/Lex/Lexer.cpp
index 34b16c7477..ea2a2deb0f 100644
--- a/lib/Lex/Lexer.cpp
+++ b/lib/Lex/Lexer.cpp
@@ -71,9 +71,22 @@ void Lexer::InitLexer(const char *BufStart, const char *BufPtr,
          "We assume that the input buffer has a null character at the end"
          " to simplify lexing!");
 
+  // Check whether we have a BOM in the beginning of the buffer. If yes - act
+  // accordingly. Right now we support only UTF-8 with and without BOM, so, just
+  // skip the UTF-8 BOM if it's present.
+  if (BufferStart == BufferPtr) {
+    // Determine the size of the BOM.
+    size_t BOMLength = llvm::StringSwitch<size_t>(BufferStart)
+      .StartsWith("\xEF\xBB\xBF", 3) // UTF-8 BOM
+      .Default(0);
+
+    // Skip the BOM.
+    BufferPtr += BOMLength;
+  }
+
   Is_PragmaLexer = false;
   IsInConflictMarker = false;
-  
+
   // Start of the file is a start of line.
   IsAtStartOfLine = true;