cmListFileLexer: Allow a leading UTF-8 Byte-Order-Mark (#11137)

Teach the lexer to read a UTF-8, UTF-16 BE/LE, or UTF-32 BE/LE
Byte-Order-Mark from the start of a file if any is present.  Report an
error on files using UTF-16 or UTF-32 and accept a UTF-8 or missing BOM.

4 files changed, 133 insertions, 10 deletions

diff --git a/Source/cmListFileCache.cxx b/Source/cmListFileCache.cxx
index 898f379..f6ea4b1 100644
--- a/Source/cmListFileCache.cxx
+++ b/Source/cmListFileCache.cxx
@@ -57,13 +57,26 @@ cmListFileParser::~cmListFileParser()
 bool cmListFileParser::ParseFile()
 {
   // Open the file.
-  if(!cmListFileLexer_SetFileName(this->Lexer, this->FileName))
+  cmListFileLexer_BOM bom;
+  if(!cmListFileLexer_SetFileName(this->Lexer, this->FileName, &bom))
     {
     cmSystemTools::Error("cmListFileCache: error can not open file ",
                          this->FileName);
     return false;
     }
 
+  // Verify the Byte-Order-Mark, if any.
+  if(bom != cmListFileLexer_BOM_None &&
+     bom != cmListFileLexer_BOM_UTF8)
+    {
+    cmListFileLexer_SetFileName(this->Lexer, 0, 0);
+    cmOStringStream m;
+    m << "File\n  " << this->FileName << "\n"
+      << "starts with a Byte-Order-Mark that is not UTF-8.";
+    this->Makefile->IssueMessage(cmake::FATAL_ERROR, m.str());
+    return false;
+    }
+
   // Use a simple recursive-descent parser to process the token
   // stream.
   bool haveNewline = true;
diff --git a/Source/cmListFileLexer.c b/Source/cmListFileLexer.c
index ad5a83d..394bd17 100644
--- a/Source/cmListFileLexer.c
+++ b/Source/cmListFileLexer.c
@@ -2307,19 +2307,68 @@ cmListFileLexer* cmListFileLexer_New()
 /*--------------------------------------------------------------------------*/
 void cmListFileLexer_Delete(cmListFileLexer* lexer)
 {
-  cmListFileLexer_SetFileName(lexer, 0);
+  cmListFileLexer_SetFileName(lexer, 0, 0);
   free(lexer);
 }
 
 /*--------------------------------------------------------------------------*/
-int cmListFileLexer_SetFileName(cmListFileLexer* lexer, const char* name)
+static cmListFileLexer_BOM cmListFileLexer_ReadBOM(FILE* f)
+{
+  unsigned char b[2];
+  if(fread(b, 1, 2, f) == 2)
+    {
+    if(b[0] == 0xEF && b[1] == 0xBB)
+      {
+      if(fread(b, 1, 1, f) == 1 && b[0] == 0xBF)
+        {
+        return cmListFileLexer_BOM_UTF8;
+        }
+      }
+    else if(b[0] == 0xFE && b[1] == 0xFF)
+      {
+      /* UTF-16 BE */
+      return cmListFileLexer_BOM_UTF16BE;
+      }
+    else if(b[0] == 0 && b[1] == 0)
+      {
+      if(fread(b, 1, 2, f) == 2 && b[0] == 0xFE && b[1] == 0xFF)
+        {
+        return cmListFileLexer_BOM_UTF32BE;
+        }
+      }
+    else if(b[0] == 0xFF && b[1] == 0xFE)
+      {
+      fpos_t p;
+      fgetpos(f, &p);
+      if(fread(b, 1, 2, f) == 2 && b[0] == 0 && b[1] == 0)
+        {
+        return cmListFileLexer_BOM_UTF32LE;
+        }
+      fsetpos(f, &p);
+      return cmListFileLexer_BOM_UTF16LE;
+      }
+    }
+  rewind(f);
+  return cmListFileLexer_BOM_None;
+}
+
+/*--------------------------------------------------------------------------*/
+int cmListFileLexer_SetFileName(cmListFileLexer* lexer, const char* name,
+                                cmListFileLexer_BOM* bom)
 {
   int result = 1;
   cmListFileLexerDestroy(lexer);
   if(name)
     {
     lexer->file = fopen(name, "r");
-    if(!lexer->file)
+    if(lexer->file)
+      {
+      if(bom)
+        {
+        *bom = cmListFileLexer_ReadBOM(lexer->file);
+        }
+      }
+    else
       {
       result = 0;
       }
@@ -2365,7 +2414,7 @@ cmListFileLexer_Token* cmListFileLexer_Scan(cmListFileLexer* lexer)
     }
   else
     {
-    cmListFileLexer_SetFileName(lexer, 0);
+    cmListFileLexer_SetFileName(lexer, 0, 0);
     return 0;
     }
 }
diff --git a/Source/cmListFileLexer.h b/Source/cmListFileLexer.h
index cc78b5c..719347c 100644
--- a/Source/cmListFileLexer.h
+++ b/Source/cmListFileLexer.h
@@ -36,6 +36,17 @@ struct cmListFileLexer_Token_s
   int column;
 };
 
+enum cmListFileLexer_BOM_e
+{
+  cmListFileLexer_BOM_None,
+  cmListFileLexer_BOM_UTF8,
+  cmListFileLexer_BOM_UTF16BE,
+  cmListFileLexer_BOM_UTF16LE,
+  cmListFileLexer_BOM_UTF32BE,
+  cmListFileLexer_BOM_UTF32LE
+};
+typedef enum cmListFileLexer_BOM_e cmListFileLexer_BOM;
+
 typedef struct cmListFileLexer_s cmListFileLexer;
 
 #ifdef __cplusplus
@@ -44,7 +55,8 @@ extern "C"
 #endif
 
 cmListFileLexer* cmListFileLexer_New();
-int cmListFileLexer_SetFileName(cmListFileLexer*, const char*);
+int cmListFileLexer_SetFileName(cmListFileLexer*, const char*,
+                                cmListFileLexer_BOM* bom);
 int cmListFileLexer_SetString(cmListFileLexer*, const char*);
 cmListFileLexer_Token* cmListFileLexer_Scan(cmListFileLexer*);
 long cmListFileLexer_GetCurrentLine(cmListFileLexer*);
diff --git a/Source/cmListFileLexer.in.l b/Source/cmListFileLexer.in.l
index 89f2917..a660d37 100644
--- a/Source/cmListFileLexer.in.l
+++ b/Source/cmListFileLexer.in.l
@@ -328,19 +328,68 @@ cmListFileLexer* cmListFileLexer_New()
 /*--------------------------------------------------------------------------*/
 void cmListFileLexer_Delete(cmListFileLexer* lexer)
 {
-  cmListFileLexer_SetFileName(lexer, 0);
+  cmListFileLexer_SetFileName(lexer, 0, 0);
   free(lexer);
 }
 
 /*--------------------------------------------------------------------------*/
-int cmListFileLexer_SetFileName(cmListFileLexer* lexer, const char* name)
+static cmListFileLexer_BOM cmListFileLexer_ReadBOM(FILE* f)
+{
+  unsigned char b[2];
+  if(fread(b, 1, 2, f) == 2)
+    {
+    if(b[0] == 0xEF && b[1] == 0xBB)
+      {
+      if(fread(b, 1, 1, f) == 1 && b[0] == 0xBF)
+        {
+        return cmListFileLexer_BOM_UTF8;
+        }
+      }
+    else if(b[0] == 0xFE && b[1] == 0xFF)
+      {
+      /* UTF-16 BE */
+      return cmListFileLexer_BOM_UTF16BE;
+      }
+    else if(b[0] == 0 && b[1] == 0)
+      {
+      if(fread(b, 1, 2, f) == 2 && b[0] == 0xFE && b[1] == 0xFF)
+        {
+        return cmListFileLexer_BOM_UTF32BE;
+        }
+      }
+    else if(b[0] == 0xFF && b[1] == 0xFE)
+      {
+      fpos_t p;
+      fgetpos(f, &p);
+      if(fread(b, 1, 2, f) == 2 && b[0] == 0 && b[1] == 0)
+        {
+        return cmListFileLexer_BOM_UTF32LE;
+        }
+      fsetpos(f, &p);
+      return cmListFileLexer_BOM_UTF16LE;
+      }
+    }
+  rewind(f);
+  return cmListFileLexer_BOM_None;
+}
+
+/*--------------------------------------------------------------------------*/
+int cmListFileLexer_SetFileName(cmListFileLexer* lexer, const char* name,
+                                cmListFileLexer_BOM* bom)
 {
   int result = 1;
   cmListFileLexerDestroy(lexer);
   if(name)
     {
     lexer->file = fopen(name, "r");
-    if(!lexer->file)
+    if(lexer->file)
+      {
+      if(bom)
+        {
+        *bom = cmListFileLexer_ReadBOM(lexer->file);
+        }
+      }
+    else
       {
       result = 0;
       }
@@ -386,7 +435,7 @@ cmListFileLexer_Token* cmListFileLexer_Scan(cmListFileLexer* lexer)
     }
   else
     {
-    cmListFileLexer_SetFileName(lexer, 0);
+    cmListFileLexer_SetFileName(lexer, 0, 0);
     return 0;
     }
 }