Remove Utf8InputBuffer

src/unicode.h

 // Copyright 2011 the V8 project authors. All rights reserved.
 // Redistribution and use in source and binary forms, with or without
 // modification, are permitted provided that the following conditions are
 // met:
 //
 //     * Redistributions of source code must retain the above copyright
 //       notice, this list of conditions and the following disclaimer.
 //     * Redistributions in binary form must reproduce the above
 //       copyright notice, this list of conditions and the following
 //       disclaimer in the documentation and/or other materials provided
@@ skipping 11 matching lines @@
 // LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
 // DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
 // THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
 // (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
 // OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 
 #ifndef V8_UNICODE_H_
 #define V8_UNICODE_H_
 
 #include <sys/types.h>
-
+#include <stdint.h>
+#include <globals.h>
 /**
  * \file
  * Definitions and convenience functions for working with unicode.
  */
 
 namespace unibrow {
 
 typedef unsigned int uchar;
 typedef unsigned char byte;
 
@@ skipping 90 matching lines @@
   static const uchar kMaxNonSurrogateCharCode = 0xffff;
   // Encoding a single UTF-16 code unit will produce 1, 2 or 3 bytes
   // of UTF-8 data.  The special case where the unit is a surrogate
   // trail produces 1 byte net, because the encoding of the pair is
   // 4 bytes and the 3 bytes that were used to encode the lead surrogate
   // can be reclaimed.
   static const int kMaxExtraUtf8BytesForOneUtf16CodeUnit = 3;
   // One UTF-16 surrogate is endoded (illegally) as 3 UTF-8 bytes.
   // The illegality stems from the surrogate not being part of a pair.
   static const int kUtf8BytesToCodeASurrogate = 3;
-  static inline uchar LeadSurrogate(int char_code) {
+  static inline uint16_t LeadSurrogate(uint32_t char_code) {
     return 0xd800 + (((char_code - 0x10000) >> 10) & 0x3ff);
   }
-  static inline uchar TrailSurrogate(int char_code) {
+  static inline uint16_t TrailSurrogate(uint32_t char_code) {
     return 0xdc00 + (char_code & 0x3ff);
   }
 };
 
 
 class Utf8 {
  public:
   static inline uchar Length(uchar chr, int previous);
   static inline unsigned Encode(
       char* out, uchar c, int previous);
-  static const byte* ReadBlock(Buffer<const char*> str, byte* buffer,
-      unsigned capacity, unsigned* chars_read, unsigned* offset);
   static uchar CalculateValue(const byte* str,
                               unsigned length,
                               unsigned* cursor);
   static const uchar kBadChar = 0xFFFD;
   static const unsigned kMaxEncodedSize   = 4;
   static const unsigned kMaxOneByteChar   = 0x7f;
   static const unsigned kMaxTwoByteChar   = 0x7ff;
   static const unsigned kMaxThreeByteChar = 0xffff;
   static const unsigned kMaxFourByteChar  = 0x1fffff;
 
@@ skipping 65 matching lines @@
   unsigned offset_;
   // The input string
   Input input_;
   // To avoid heap allocation, we keep an internal buffer to which
   // the encoded string can write its characters.  The string
   // implementation is free to decide whether it wants to use this
   // buffer or not.
   byte util_buffer_[kSize];
 };
 
-// --- U t f 8   I n p u t   B u f f e r ---
 
-template <unsigned s = 256>
-class Utf8InputBuffer : public InputBuffer<Utf8, Buffer<const char*>, s> {
+class Utf8DecoderBase {
  public:
-  inline Utf8InputBuffer() { }
-  inline Utf8InputBuffer(const char* data, unsigned length);
-  inline void Reset(const char* data, unsigned length) {
-    InputBuffer<Utf8, Buffer<const char*>, s>::Reset(
-        Buffer<const char*>(data, length));
-  }
+  // Initialization done in subclass.
+  inline Utf8DecoderBase();
+  inline Utf8DecoderBase(uint16_t* buffer,
+                         unsigned buffer_length,
+                         const uint8_t* stream,
+                         unsigned stream_length);
+  inline unsigned Utf16Length() const { return utf16_length_; }
+ protected:
+  // This reads all characters and sets the utf16_length_.
+  // The first buffer_length utf16 chars are cached in the buffer.
+  void Reset(uint16_t* buffer,
+             unsigned buffer_length,
+             const uint8_t* stream,
+             unsigned stream_length);
+  static void WriteUtf16Slow(const uint8_t* stream,
+                             uint16_t* data,
+                             unsigned length);
+  const uint8_t* unbuffered_start_;
+  unsigned utf16_length_;
+  bool last_byte_of_buffer_unused_;
+ private:
+  DISALLOW_COPY_AND_ASSIGN(Utf8DecoderBase);
+};
+
+template <unsigned kBufferSize>
+class Utf8Decoder : public Utf8DecoderBase {
+ public:
+  inline Utf8Decoder() {}
+  inline Utf8Decoder(const char* stream, unsigned length);
+  inline void Reset(const char* stream, unsigned length);
+  inline unsigned WriteUtf16(uint16_t* data, unsigned length) const;
+ private:
+  uint16_t buffer_[kBufferSize];
 };
 
 
 struct Uppercase {
   static bool Is(uchar c);
 };
 struct Lowercase {
   static bool Is(uchar c);
 };
 struct Letter {
@@ skipping 49 matching lines @@
   static const int kMaxWidth = 1;
   static int Convert(uchar c,
                      uchar n,
                      uchar* result,
                      bool* allow_caching_ptr);
 };
 
 }  // namespace unibrow
 
 #endif  // V8_UNICODE_H_