Add U0 case to Utf8ToUtf32

[platform/core/uifw/dali-toolkit.git] / dali-toolkit / internal / text / character-set-conversion.cpp

diff --git a/dali-toolkit/internal/text/character-set-conversion.cpp b/dali-toolkit/internal/text/character-set-conversion.cpp
index abda102..d79fbc3 100644 (file)
--- a/dali-toolkit/internal/text/character-set-conversion.cpp
+++ b/dali-toolkit/internal/text/character-set-conversion.cpp
@@ -24,6 +24,9 @@ namespace Dali
 namespace Toolkit
 {
 
+namespace Text
+{
+
 namespace
 {
   const static uint8_t U1 = 1u;
@@ -66,8 +69,16 @@ namespace
     U0, U0, U0, U0,                         // Non valid.
     U0, U0, U0, U0,                         // Non valid.
   };
+
+  const uint8_t CR = 0xd;
+  const uint8_t LF = 0xa;
 } // namespace
 
+uint8_t GetUtf8Length( uint8_t utf8LeadByte )
+{
+  return UTF8_LENGTH[utf8LeadByte];
+}
+
 uint32_t GetNumberOfUtf8Characters( const uint8_t* const utf8, uint32_t length )
 {
   uint32_t numberOfCharacters = 0u;
@@ -127,8 +138,26 @@ uint32_t Utf8ToUtf32( const uint8_t* const utf8, uint32_t length, uint32_t* utf3
     {
       case U1:
       {
-        *utf32++ = leadByte;
-        begin++;
+        if( CR == leadByte )
+        {
+          // Replace CR+LF or CR by LF
+          *utf32++ = LF;
+
+          // Look ahead if the next one is a LF.
+          ++begin;
+          if( begin < end )
+          {
+            if( LF == *begin )
+            {
+              ++begin;
+            }
+          }
+        }
+        else
+        {
+          *utf32++ = leadByte;
+          begin++;
+        }
         break;
       }
 
@@ -145,7 +174,7 @@ uint32_t Utf8ToUtf32( const uint8_t* const utf8, uint32_t length, uint32_t* utf3
       case U3:
       {
         uint32_t& code = *utf32++;
-        code = leadByte & 0x1fu;
+        code = leadByte & 0x0fu;
         begin++;
         code <<= 6u;
         code |= *begin++ & 0x3fu;
@@ -157,7 +186,7 @@ uint32_t Utf8ToUtf32( const uint8_t* const utf8, uint32_t length, uint32_t* utf3
       case U4:
       {
         uint32_t& code = *utf32++;
-        code = leadByte & 0x1fu;
+        code = leadByte & 0x07u;
         begin++;
         code <<= 6u;
         code |= *begin++ & 0x3fu;
@@ -167,17 +196,26 @@ uint32_t Utf8ToUtf32( const uint8_t* const utf8, uint32_t length, uint32_t* utf3
         code |= *begin++ & 0x3fu;
         break;
       }
+
+      case U0:    // Invalid case
+      {
+        begin++;
+        *utf32++ = 0x20;    // Use white space
+        break;
+      }
     }
   }
 
   return numberOfCharacters;
 }
 
-void Utf32ToUtf8( const uint32_t* const utf32, uint32_t numberOfCharacters, uint8_t* utf8 )
+uint32_t Utf32ToUtf8( const uint32_t* const utf32, uint32_t numberOfCharacters, uint8_t* utf8 )
 {
   const uint32_t* begin = utf32;
   const uint32_t* end = utf32 + numberOfCharacters;
 
+  uint8_t* utf8Begin = utf8;
+
   for( ; begin < end; ++begin )
   {
     const uint32_t code = *begin;
@@ -205,8 +243,23 @@ void Utf32ToUtf8( const uint32_t* const utf32, uint32_t numberOfCharacters, uint
       *utf8++ = static_cast<uint8_t>(   code          & 0x3f ) | 0x80u; // continuation byte
     }
   }
+
+  return utf8 - utf8Begin;
 }
 
+void Utf32ToUtf8( const uint32_t* const utf32, uint32_t numberOfCharacters, std::string& utf8 )
+{
+  utf8.clear();
+
+  uint32_t numberOfBytes = GetNumberOfUtf8Bytes( &utf32[0], numberOfCharacters );
+  utf8.resize( numberOfBytes );
+
+  // This is a bit horrible but std::string returns a (signed) char*
+  Utf32ToUtf8( utf32, numberOfCharacters, reinterpret_cast<uint8_t*>(&utf8[0]) );
+}
+
+} // namespace Text
+
 } // namespace Toolkit
 
 } // namespace Dali