Notepad++공개소스 안에 포함된 UTF8, UTF16에 대한 소스

Window API함수로는 다른 한 줄에 해결되는 것이 있겠지만, 실제 제대로 이해하고 순수 C/C++로 구현하면 아래와 같은 코드가 나오는 군요. Cross Platform을 지원하는 소스에서 많이 배우고 있습니다. 기본적인 프로그래밍적 사고부터 말이죠. GNU 왠지 멋져 보임.

// Scintilla source code edit control
/** @file UniConversion.cxx
 ** Functions to handle UTF-8 and UTF-16 strings.
 **/
// Copyright 1998-2001 by Neil Hodgson <neilh@scintilla.org>
// The License.txt file describes the conditions under which this software may be distributed.

#include <stdlib.h>

#include "UniConversion.h"

enum { SURROGATE_LEAD_FIRST = 0xD800 };
enum { SURROGATE_TRAIL_FIRST = 0xDC00 };
enum { SURROGATE_TRAIL_LAST = 0xDFFF };

unsigned int UTF8Length(const wchar_t *uptr, unsigned int tlen) {
 unsigned int len = 0;
 for (unsigned int i = 0; i < tlen && uptr[i];) {
  unsigned int uch = uptr[i];
  if (uch < 0x80) {
   len++;
  } else if (uch < 0x800) {
   len += 2;
  } else if ((uch >= SURROGATE_LEAD_FIRST) &&
   (uch <= SURROGATE_TRAIL_LAST)) {
   len += 4;
   i++;
  } else {
   len += 3;
  }
  i++;
 }
 return len;
}

void UTF8FromUTF16(const wchar_t *uptr, unsigned int tlen, char *putf, unsigned int len) {
 int k = 0;
 for (unsigned int i = 0; i < tlen && uptr[i];) {
  unsigned int uch = uptr[i];
  if (uch < 0x80) {
   putf[k++] = static_cast<char>(uch);
  } else if (uch < 0x800) {
   putf[k++] = static_cast<char>(0xC0 | (uch >> 6));
   putf[k++] = static_cast<char>(0x80 | (uch & 0x3f));
  } else if ((uch >= SURROGATE_LEAD_FIRST) &&
   (uch <= SURROGATE_TRAIL_LAST)) {
   // Half a surrogate pair
   i++;
   unsigned int xch = 0x10000 + ((uch & 0x3ff) << 10) + (uptr[i] & 0x3ff);
   putf[k++] = static_cast<char>(0xF0 | (xch >> 18));
   putf[k++] = static_cast<char>(0x80 | ((xch >> 12) & 0x3f));
   putf[k++] = static_cast<char>(0x80 | ((xch >> 6) & 0x3f));
   putf[k++] = static_cast<char>(0x80 | (xch & 0x3f));
  } else {
   putf[k++] = static_cast<char>(0xE0 | (uch >> 12));
   putf[k++] = static_cast<char>(0x80 | ((uch >> 6) & 0x3f));
   putf[k++] = static_cast<char>(0x80 | (uch & 0x3f));
  }
  i++;
 }
 putf[len] = '\0';
}

unsigned int UTF8CharLength(unsigned char ch) {
 if (ch < 0x80) {
  return 1;
 } else if (ch < 0x80 + 0x40 + 0x20) {
  return 2;
 } else if (ch < 0x80 + 0x40 + 0x20 + 0x10) {
  return 3;
 } else {
  return 4;
 }
}

unsigned int UTF16Length(const char *s, unsigned int len) {
 unsigned int ulen = 0;
 unsigned int charLen;
 for (unsigned int i=0; i<len;) {
  unsigned char ch = static_cast<unsigned char>(s[i]);
  if (ch < 0x80) {
   charLen = 1;
  } else if (ch < 0x80 + 0x40 + 0x20) {
   charLen = 2;
  } else if (ch < 0x80 + 0x40 + 0x20 + 0x10) {
   charLen = 3;
  } else {
   charLen = 4;
   ulen++;
  }
  i += charLen;
  ulen++;
 }
 return ulen;
}

unsigned int UTF16FromUTF8(const char *s, unsigned int len, wchar_t *tbuf, unsigned int tlen) {
 unsigned int ui=0;
 const unsigned char *us = reinterpret_cast<const unsigned char *>(s);
 unsigned int i=0;
 while ((i<len) && (ui<tlen)) {
  unsigned char ch = us[i++];
  if (ch < 0x80) {
   tbuf[ui] = ch;
  } else if (ch < 0x80 + 0x40 + 0x20) {
   tbuf[ui] = static_cast<wchar_t>((ch & 0x1F) << 6);
   ch = us[i++];
   tbuf[ui] = static_cast<wchar_t>(tbuf[ui] + (ch & 0x7F));
  } else if (ch < 0x80 + 0x40 + 0x20 + 0x10) {
   tbuf[ui] = static_cast<wchar_t>((ch & 0xF) << 12);
   ch = us[i++];
   tbuf[ui] = static_cast<wchar_t>(tbuf[ui] + ((ch & 0x7F) << 6));
   ch = us[i++];
   tbuf[ui] = static_cast<wchar_t>(tbuf[ui] + (ch & 0x7F));
  } else {
   // Outside the BMP so need two surrogates
   int val = (ch & 0x7) << 18;
   ch = us[i++];
   val += (ch & 0x3F) << 12;
   ch = us[i++];
   val += (ch & 0x3F) << 6;
   ch = us[i++];
   val += (ch & 0x3F);
   tbuf[ui] = static_cast<wchar_t>(((val - 0x10000) >> 10) + SURROGATE_LEAD_FIRST);
   ui++;
   tbuf[ui] = static_cast<wchar_t>((val & 0x3ff) + SURROGATE_TRAIL_FIRST);
  }
  ui++;
 }
 return ui;
}