Add support for Unicode rendering (#47)

This uses utfcpp combined with a custom font, in the form of a PNG and text file. By default, the game acts exactly as it did before; custom fonts can be provided by third parties.

desktop_version/CMakeLists.txt

@@ -56,6 +56,7 @@ INCLUDE_DIRECTORIES(
 	../third_party/tinyxml
 	../third_party/physfs
 	../third_party/lodepng
+	../third_party/utfcpp/source
 )
 
 # Source Lists

desktop_version/src/FileSystemUtils.cpp

@@ -179,7 +179,7 @@ bool FILESYSTEM_loadTiXmlDocument(const char *name, TiXmlDocument *doc)
 	{
 		return false;
 	}
-	doc->Parse((const char*)mem);
+	doc->Parse((const char*)mem, NULL, TIXML_ENCODING_UTF8);
 	FILESYSTEM_freeMemory(&mem);
 	return true;
 }

desktop_version/src/Graphics.cpp

@@ -3,6 +3,10 @@
 #include "Entity.h"
 #include "Map.h"
 #include "Screen.h"
+#include "FileSystemUtils.h"
+#include <utf8/unchecked.h>
+#include <stdio.h>
+#include <stdlib.h>
 
 void Graphics::init()
 {
@@ -129,6 +133,22 @@ void Graphics::init()
     showmousecursor = true;
 }
 
+int Graphics::font_idx(char32_t ch) {
+    if (font_positions.size() > 0) {
+        std::map<int, int>::iterator iter = font_positions.find(ch);
+        if (iter == font_positions.end()) {
+            iter = font_positions.find('?');
+            if (iter == font_positions.end()) {
+                puts("font.txt missing fallback character!");
+                exit(1);
+            }
+        }
+        return iter->second;
+    } else {
+        return ch;
+    }
+}
+
 Graphics::~Graphics()
 {
 
@@ -146,7 +166,7 @@ void Graphics::drawspritesetcol(int x, int y, int t, int c, UtilityClass& help)
 
 void Graphics::Makebfont()
 {
-    for (int j =  0; j < 16; j++)
+    for (int j =  0; j < (grphx.im_bfont->h / 8); j++)
     {
         for (int i = 0; i < 16; i++)
         {
@@ -154,25 +174,31 @@ void Graphics::Makebfont()
             SDL_Surface* temp = GetSubSurface(grphx.im_bfont,i*8,j*8,8,8);
             bfont.push_back(temp);
 
-            temp = GetSubSurface(grphx.im_bfont,i*8,j*8,8,8);
             SDL_Surface* TempFlipped = FlipSurfaceVerticle(temp);
-
             flipbfont.push_back(TempFlipped);
-			SDL_FreeSurface(temp);
-
         }
     }
 
-    //Ok, now we work out the lengths (this data string cortesy of a program I wrote!)
+    unsigned char* charmap = NULL;
-    for (int i = 0; i < 256; i++)
+    size_t length;
-    {
+    FILESYSTEM_loadFileToMemory("graphics/font.txt", &charmap, &length);
-        bfontlen.push_back(6);
+    if (charmap != NULL) {
+        unsigned char* current = charmap;
+        unsigned char* end = charmap + length;
+        int pos = 0;
+        while (current != end) {
+            int codepoint = utf8::unchecked::next(current);
+            font_positions[codepoint] = pos;
+            ++pos;
+        }
+    }
 }
 
-
+int Graphics::bfontlen(char32_t ch) {
-    for(int k = 0; k < 96; k++)
+    if (ch < 32) {
-    {
+        return 6;
-        bfontlen[k + 32] = 8;// int(maprow[k]);
+    } else {
+        return 8;
     }
 }
 
@@ -250,13 +276,9 @@ void Graphics::Print( int _x, int _y, std::string _s, int r, int g, int b, bool
         _x = ((160 ) - ((len(_s)) / 2));
     int bfontpos = 0;
     int curr;
-    for (unsigned int i = 0; i < _s.length(); i++)
+    std::string::iterator iter = _s.begin();
-    {
+    while (iter != _s.end()) {
-        curr = (_s.c_str())[i];
+        curr = utf8::unchecked::next(iter);
-        if (curr > 255 || curr < 0)
-        {
-            curr = '?';
-        }
         point tpoint;
         tpoint.x = _x + bfontpos;
         tpoint.y = _y;
@@ -267,13 +289,13 @@ void Graphics::Print( int _x, int _y, std::string _s, int r, int g, int b, bool
 
         if (flipmode)
         {
-            BlitSurfaceColoured( flipbfont[curr], NULL, backBuffer, &fontRect , ct);
+            BlitSurfaceColoured( flipbfont[font_idx(curr)], NULL, backBuffer, &fontRect , ct);
         }
         else
         {
-            BlitSurfaceColoured( bfont[curr], NULL, backBuffer, &fontRect , ct);
+            BlitSurfaceColoured( bfont[font_idx(curr)], NULL, backBuffer, &fontRect , ct);
         }
-        bfontpos+=bfontlen[curr] ;
+        bfontpos+=bfontlen(curr) ;
     }
 }
 
@@ -293,13 +315,9 @@ void Graphics::bigprint(  int _x, int _y, std::string _s, int r, int g, int b, b
 
     int bfontpos = 0;
     int curr;
-    for (unsigned int i = 0; i < _s.length(); i++)
+    std::string::iterator iter = _s.begin();
-    {
+    while (iter != _s.end()) {
-        curr = (_s.c_str())[i];
+        curr = utf8::unchecked::next(iter);
-        if (curr > 255 || curr < 0)
-        {
-            curr = '?';
-        }
 
         /*
         point tpoint;
@@ -313,29 +331,29 @@ void Graphics::bigprint(  int _x, int _y, std::string _s, int r, int g, int b, b
 
         if (flipmode)
         {
-			SDL_Surface* tempPrint = ScaleSurfaceSlow(flipbfont[curr], bfont[curr]->w *sc,bfont[curr]->h *sc);
+			SDL_Surface* tempPrint = ScaleSurfaceSlow(flipbfont[font_idx(curr)], bfont[font_idx(curr)]->w *sc,bfont[font_idx(curr)]->h *sc);
 			SDL_Rect printrect = { Sint16((_x) + bfontpos), Sint16(_y) , Sint16(bfont_rect.w*sc), Sint16(bfont_rect.h * sc)};
 			BlitSurfaceColoured(tempPrint, NULL, backBuffer, &printrect, ct);
 			SDL_FreeSurface(tempPrint);
         }
         else
         {
-			SDL_Surface* tempPrint = ScaleSurfaceSlow(bfont[curr], bfont[curr]->w *sc,bfont[curr]->h *sc);
+			SDL_Surface* tempPrint = ScaleSurfaceSlow(bfont[font_idx(curr)], bfont[font_idx(curr)]->w *sc,bfont[font_idx(curr)]->h *sc);
 			SDL_Rect printrect = { static_cast<Sint16>((_x) + bfontpos), static_cast<Sint16>(_y) , static_cast<Sint16>((bfont_rect.w*sc)+1), static_cast<Sint16>((bfont_rect.h * sc)+1)};
 			BlitSurfaceColoured(tempPrint, NULL, backBuffer, &printrect, ct);
 			SDL_FreeSurface(tempPrint);
         }
-        bfontpos+=bfontlen[curr] *sc;
+        bfontpos+=bfontlen(curr) *sc;
     }
 }
 
 int Graphics::len(std::string t)
 {
     int bfontpos = 0;
-    for (unsigned int i = 0; i < t.length(); i++)
+    std::string::iterator iter = t.begin();
-    {
+    while (iter != t.end()) {
-        int cur = (t.c_str())[i];
+        int cur = utf8::unchecked::next(iter);
-        bfontpos+= bfontlen[cur] ;
+        bfontpos += bfontlen(cur);
     }
     return bfontpos;
 }
@@ -351,14 +369,9 @@ void Graphics::PrintOff( int _x, int _y, std::string _s, int r, int g, int b, bo
     if (cen)
         _x = ((160) - (len(_s) / 2))+_x;
     int bfontpos = 0;
-    int curr;
+    std::string::iterator iter = _s.begin();
-    for (unsigned int i = 0; i < _s.length(); i++)
+    while (iter != _s.end()) {
-    {
+        int curr = utf8::unchecked::next(iter);
-        curr = (_s.c_str())[i];
-        if (curr > 255 || curr < 0)
-        {
-            curr = '?';
-        }
         point tpoint;
         tpoint.x = _x + bfontpos;
         tpoint.y = _y;
@@ -369,16 +382,16 @@ void Graphics::PrintOff( int _x, int _y, std::string _s, int r, int g, int b, bo
 
         if (flipmode)
         {
-            //flipbfont[cur].colorTransform(bfont_rect, ct);
+            //flipbfont[font_idx(cur)].colorTransform(bfont_rect, ct);
-            BlitSurfaceColoured( bfont[curr], NULL, backBuffer, &fontRect , ct);
+            BlitSurfaceColoured( bfont[font_idx(curr)], NULL, backBuffer, &fontRect , ct);
         }
         else
         {
-            //bfont[cur].colorTransform(bfont_rect, ct);
+            //bfont[font_idx(cur)].colorTransform(bfont_rect, ct);
-            //backBuffer.copyPixels(bfont[cur], bfont_rect, tpoint);
+            //backBuffer.copyPixels(bfont[font_idx(cur)], bfont_rect, tpoint);
-            BlitSurfaceColoured( bfont[curr], NULL, backBuffer, &fontRect , ct);
+            BlitSurfaceColoured( bfont[font_idx(curr)], NULL, backBuffer, &fontRect , ct);
         }
-        bfontpos+=bfontlen[curr] ;
+        bfontpos+=bfontlen(curr) ;
     }
 }
 
@@ -417,13 +430,9 @@ void Graphics::RPrint( int _x, int _y, std::string _s, int r, int g, int b, bool
         _x = ((308) - (_s.length() / 2));
     int bfontpos = 0;
     int curr;
-    for (unsigned int i = 0; i < _s.length(); i++)
+    std::string::iterator iter = _s.begin();
-    {
+    while (iter != _s.end()) {
-        curr = (_s.c_str())[i];
+        curr = utf8::unchecked::next(iter);
-        if (curr > 255 || curr < 0)
-        {
-            curr = '?';
-        }
         point tpoint;
         tpoint.x = _x + bfontpos;
         tpoint.y = _y;
@@ -434,16 +443,16 @@ void Graphics::RPrint( int _x, int _y, std::string _s, int r, int g, int b, bool
 
         if (flipmode)
         {
-            //flipbfont[cur].colorTransform(bfont_rect, ct);
+            //flipbfont[font_idx(cur)].colorTransform(bfont_rect, ct);
-            BlitSurfaceColoured( flipbfont[curr], NULL, backBuffer, &fontRect , ct);
+            BlitSurfaceColoured( flipbfont[font_idx(curr)], NULL, backBuffer, &fontRect , ct);
         }
         else
         {
-            //bfont[cur].colorTransform(bfont_rect, ct);
+            //bfont[font_idx(cur)].colorTransform(bfont_rect, ct);
-            //backBuffer.copyPixels(bfont[cur], bfont_rect, tpoint);
+            //backBuffer.copyPixels(bfont[font_idx(cur)], bfont_rect, tpoint);
-            BlitSurfaceColoured( bfont[curr], NULL, backBuffer, &fontRect , ct);
+            BlitSurfaceColoured( bfont[font_idx(curr)], NULL, backBuffer, &fontRect , ct);
         }
-        bfontpos+=bfontlen[curr] ;
+        bfontpos+=bfontlen(curr) ;
     }
 }
 
@@ -1035,7 +1044,8 @@ void Graphics::createtextbox( std::string t, int xp, int yp, int r/*= 255*/, int
         textbox[m].clear();
         textbox[m].line[0] = t;
         textbox[m].xp = xp;
-        if (xp == -1) textbox[m].xp = 160 - (((t.length() / 2) + 1) * 8);
+        int length = utf8::unchecked::distance(t.begin(), t.end());
+        if (xp == -1) textbox[m].xp = 160 - (((length / 2) + 1) * 8);
         textbox[m].yp = yp;
         textbox[m].initcol(r, g, b);
         textbox[m].resize();
@@ -3100,28 +3110,24 @@ void Graphics::bigrprint(int x, int y, std::string& t, int r, int g, int b, bool
 
 	int bfontpos = 0;
 	int cur;
-	for (size_t i = 0; i < t.length(); i++)
+	std::string::iterator iter = t.begin();
-	{
+	while (iter != t.end()) {
-		cur = (t.c_str())[i];
+		cur = utf8::unchecked::next(iter);
-		if (cur > 255 || cur < 0)
-		{
-			cur = '?';
-		}
 		if (flipmode)
 		{
-			SDL_Surface* tempPrint = ScaleSurfaceSlow(flipbfont[cur], bfont[cur]->w *sc,bfont[cur]->h *sc);
+			SDL_Surface* tempPrint = ScaleSurfaceSlow(flipbfont[font_idx(cur)], bfont[font_idx(cur)]->w *sc,bfont[font_idx(cur)]->h *sc);
 			SDL_Rect printrect = { Sint16(x + bfontpos), Sint16(y) , Sint16(bfont_rect.w*sc), Sint16(bfont_rect.h * sc)};
 			BlitSurfaceColoured(tempPrint, NULL, backBuffer, &printrect ,ct);
 			SDL_FreeSurface(tempPrint);
 		}
 		else
 		{
-			SDL_Surface* tempPrint = ScaleSurfaceSlow(bfont[cur], bfont[cur]->w *sc,bfont[cur]->h *sc);
+			SDL_Surface* tempPrint = ScaleSurfaceSlow(bfont[font_idx(cur)], bfont[font_idx(cur)]->w *sc,bfont[font_idx(cur)]->h *sc);
 			SDL_Rect printrect = { Sint16((x) + bfontpos), Sint16(y) , Sint16(bfont_rect.w*sc), Sint16(bfont_rect.h * sc)};
 			BlitSurfaceColoured(tempPrint, NULL, backBuffer, &printrect, ct);
 			SDL_FreeSurface(tempPrint);
 		}
-		bfontpos+=bfontlen[cur]* sc;
+		bfontpos+=bfontlen(cur)* sc;
 	}
 }
 

desktop_version/src/Graphics.h

@@ -3,6 +3,7 @@
 
 #include "GraphicsResources.h"
 #include <vector>
+#include <map>
 
 
 
@@ -28,6 +29,9 @@ public:
 
 	GraphicsResources grphx;
 
+	int bfontlen(char32_t ch);
+	int font_idx(char32_t ch);
+
 	void Makebfont();
 
 	void drawhuetile(int x, int y, int t, int c);
@@ -212,7 +216,6 @@ public:
 	std::vector <SDL_Surface*> bfontmask;
 	std::vector <SDL_Surface*> flipbfont;
 	std::vector <SDL_Surface*> flipbfontmask;
-	std::vector <int> bfontlen;
 
 	bool flipmode;
 	bool setflipmode;
@@ -280,6 +283,8 @@ public:
 	bool translucentroomname;
 
 	bool showmousecursor;
+
+	std::map<int, int> font_positions;
 };
 
 extern Graphics graphics;

desktop_version/src/KeyPoll.cpp

@@ -1,6 +1,7 @@
 #include "KeyPoll.h"
 #include <stdio.h>
 #include <string.h>
+#include <utf8/unchecked.h>
 
 void KeyPoll::setSensitivity(int _value)
 {
@@ -94,11 +95,12 @@ void KeyPoll::Poll()
 
 			if (textentrymode)
 			{
-				if (evt.key.keysym.sym == SDLK_BACKSPACE)
+				if (evt.key.keysym.sym == SDLK_BACKSPACE && !keybuffer.empty())
 				{
-					bool kbemptybefore = keybuffer.empty();
+					std::string::iterator iter = keybuffer.end();
-					keybuffer = keybuffer.substr(0, keybuffer.length() - 1);
+					utf8::unchecked::prior(iter);
-					if (!kbemptybefore && keybuffer.empty())
+					keybuffer = keybuffer.substr(0, iter - keybuffer.begin());
+					if (keybuffer.empty())
 					{
 						linealreadyemptykludge = true;
 					}

desktop_version/src/Textbox.cpp

@@ -1,4 +1,5 @@
 #include "Textbox.h"
+#include <utf8/unchecked.h>
 
 textboxclass::textboxclass()
 {
@@ -133,7 +134,8 @@ void textboxclass::resize()
     max = 0;
     for (int iter = 0; iter < numlines; iter++)
     {
-        if (line[iter].length() > (unsigned int)max) max = line[iter].length();
+        unsigned int len = utf8::unchecked::distance(line[iter].begin(), line[iter].end());
+        if (len > (unsigned int)max) max = len;
     }
 
     lw = max;

desktop_version/src/editor.cpp

@@ -16,6 +16,7 @@
 #include "FileSystemUtils.h"
 
 #include <string>
+#include <utf8/unchecked.h>
 
 edlevelclass::edlevelclass()
 {
@@ -2625,7 +2626,8 @@ void editorrender( KeyPoll& key, Graphics& dwgfx, Game& game, mapclass& map, ent
                 }
                 else
                 {
-                    fillboxabs(dwgfx, (edentity[i].x*8)- (ed.levx*40*8),(edentity[i].y*8)- (ed.levy*30*8),edentity[i].scriptname.length()*8,8,dwgfx.getRGB(96,96,96));
+                    int length = utf8::unchecked::distance(edentity[i].scriptname.begin(), edentity[i].scriptname.end());
+                    fillboxabs(dwgfx, (edentity[i].x*8)- (ed.levx*40*8),(edentity[i].y*8)- (ed.levy*30*8),length*8,8,dwgfx.getRGB(96,96,96));
                 }
                 dwgfx.bprint((edentity[i].x*8)- (ed.levx*40*8),(edentity[i].y*8)- (ed.levy*30*8), edentity[i].scriptname, 196, 196, 255 - help.glow);
                 break;
@@ -3796,7 +3798,7 @@ void editorinput( KeyPoll& key, Graphics& dwgfx, Game& game, mapclass& map, enti
                         ed.sby--;
                     }
                     key.keybuffer=ed.sb[ed.pagey+ed.sby];
-                    ed.sbx = ed.sb[ed.pagey+ed.sby].length();
+                    ed.sbx = utf8::unchecked::distance(ed.sb[ed.pagey+ed.sby].begin(), ed.sb[ed.pagey+ed.sby].end());
                 }
 
                 if (key.isDown(27))
@@ -3881,7 +3883,7 @@ void editorinput( KeyPoll& key, Graphics& dwgfx, Game& game, mapclass& map, enti
             }
 
             ed.sb[ed.pagey+ed.sby]=key.keybuffer;
-            ed.sbx = ed.sb[ed.pagey+ed.sby].length();
+            ed.sbx = utf8::unchecked::distance(ed.sb[ed.pagey+ed.sby].begin(), ed.sb[ed.pagey+ed.sby].end());
 
             if(!game.press_map && !key.isDown(27)) game.mapheld=false;
             if (!game.mapheld)
@@ -3900,7 +3902,7 @@ void editorinput( KeyPoll& key, Graphics& dwgfx, Game& game, mapclass& map, enti
                         }
                         if(ed.sby+ed.pagey>=ed.sblength) ed.sblength=ed.sby+ed.pagey;
                         key.keybuffer=ed.sb[ed.pagey+ed.sby];
-                        ed.sbx = ed.sb[ed.pagey+ed.sby].length();
+                        ed.sbx = utf8::unchecked::distance(ed.sb[ed.pagey+ed.sby].begin(), ed.sb[ed.pagey+ed.sby].end());
                     }
                     else
                     {

third_party/utfcpp/LICENSE

@@ -0,0 +1,23 @@
+Boost Software License - Version 1.0 - August 17th, 2003
+
+Permission is hereby granted, free of charge, to any person or organization
+obtaining a copy of the software and accompanying documentation covered by
+this license (the "Software") to use, reproduce, display, distribute,
+execute, and transmit the Software, and to prepare derivative works of the
+Software, and to permit third-parties to whom the Software is furnished to
+do so, all subject to the following:
+
+The copyright notices in the Software and this entire statement, including
+the above license grant, this restriction and the following disclaimer,
+must be included in all copies of the Software, in whole or in part, and
+all derivative works of the Software, unless such copies or derivative
+works are solely in the form of machine-executable object code generated by
+a source language processor.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT
+SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE
+FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE,
+ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+DEALINGS IN THE SOFTWARE.

third_party/utfcpp/source/utf8/core.h

@@ -0,0 +1,287 @@
+// Copyright 2006 Nemanja Trifunovic
+
+/*
+Permission is hereby granted, free of charge, to any person or organization
+obtaining a copy of the software and accompanying documentation covered by
+this license (the "Software") to use, reproduce, display, distribute,
+execute, and transmit the Software, and to prepare derivative works of the
+Software, and to permit third-parties to whom the Software is furnished to
+do so, all subject to the following:
+
+The copyright notices in the Software and this entire statement, including
+the above license grant, this restriction and the following disclaimer,
+must be included in all copies of the Software, in whole or in part, and
+all derivative works of the Software, unless such copies or derivative
+works are solely in the form of machine-executable object code generated by
+a source language processor.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT
+SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE
+FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE,
+ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+DEALINGS IN THE SOFTWARE.
+*/
+
+
+#ifndef UTF8_FOR_CPP_CORE_H_2675DCD0_9480_4c0c_B92A_CC14C027B731
+#define UTF8_FOR_CPP_CORE_H_2675DCD0_9480_4c0c_B92A_CC14C027B731
+
+#include <iterator>
+
+// Determine the C++ standard version.
+// If the user defines UTF_CPP_CPLUSPLUS, use that.
+// Otherwise, trust the unreliable predefined macro __cplusplus
+
+#if !defined UTF_CPP_CPLUSPLUS
+    #define UTF_CPP_CPLUSPLUS __cplusplus
+#endif
+
+#if UTF_CPP_CPLUSPLUS >= 201103L // C++ 11 or later
+    #define OVERRIDE override
+    #define NOEXCEPT noexcept
+#else // C++ 98/03
+    #define OVERRIDE
+    #define NOEXCEPT throw()
+#endif // C++ 11 or later
+
+
+namespace utf8
+{
+    // The typedefs for 8-bit, 16-bit and 32-bit unsigned integers
+    // You may need to change them to match your system.
+    // These typedefs have the same names as ones from cstdint, or boost/cstdint
+    typedef unsigned char   uint8_t;
+    typedef unsigned short  uint16_t;
+    typedef unsigned int    uint32_t;
+
+// Helper code - not intended to be directly called by the library users. May be changed at any time
+namespace internal
+{
+    // Unicode constants
+    // Leading (high) surrogates: 0xd800 - 0xdbff
+    // Trailing (low) surrogates: 0xdc00 - 0xdfff
+    const uint16_t LEAD_SURROGATE_MIN  = 0xd800u;
+    const uint16_t LEAD_SURROGATE_MAX  = 0xdbffu;
+    const uint16_t TRAIL_SURROGATE_MIN = 0xdc00u;
+    const uint16_t TRAIL_SURROGATE_MAX = 0xdfffu;
+    const uint16_t LEAD_OFFSET         = 0xd7c0u;       // LEAD_SURROGATE_MIN - (0x10000 >> 10)
+    const uint32_t SURROGATE_OFFSET    = 0xfca02400u;   // 0x10000u - (LEAD_SURROGATE_MIN << 10) - TRAIL_SURROGATE_MIN
+
+    // Maximum valid value for a Unicode code point
+    const uint32_t CODE_POINT_MAX      = 0x0010ffffu;
+
+    template<typename octet_type>
+    inline uint8_t mask8(octet_type oc)
+    {
+        return static_cast<uint8_t>(0xff & oc);
+    }
+    template<typename octet_type>
+    inline bool is_trail(octet_type oc)
+    {
+        return ((utf8::internal::mask8(oc) >> 6) == 0x2);
+    }
+
+    template <typename u16>
+    inline bool is_surrogate(u16 cp)
+    {
+        return (cp >= LEAD_SURROGATE_MIN && cp <= TRAIL_SURROGATE_MAX);
+    }
+
+    template <typename u32>
+    inline bool is_code_point_valid(u32 cp)
+    {
+        return (cp <= CODE_POINT_MAX && !utf8::internal::is_surrogate(cp));
+    }
+
+    template <typename octet_iterator>
+    inline typename std::iterator_traits<octet_iterator>::difference_type
+    sequence_length(octet_iterator lead_it)
+    {
+        uint8_t lead = utf8::internal::mask8(*lead_it);
+        if (lead < 0x80)
+            return 1;
+        else if ((lead >> 5) == 0x6)
+            return 2;
+        else if ((lead >> 4) == 0xe)
+            return 3;
+        else if ((lead >> 3) == 0x1e)
+            return 4;
+        else
+            return 0;
+    }
+
+    template <typename octet_difference_type>
+    inline bool is_overlong_sequence(uint32_t cp, octet_difference_type length)
+    {
+        if (cp < 0x80) {
+            if (length != 1) 
+                return true;
+        }
+        else if (cp < 0x800) {
+            if (length != 2) 
+                return true;
+        }
+        else if (cp < 0x10000) {
+            if (length != 3) 
+                return true;
+        }
+
+        return false;
+    }
+
+    enum utf_error {UTF8_OK, NOT_ENOUGH_ROOM, INVALID_LEAD, INCOMPLETE_SEQUENCE, OVERLONG_SEQUENCE, INVALID_CODE_POINT};
+
+    /// Helper for get_sequence_x
+    template <typename octet_iterator>
+    utf_error increase_safely(octet_iterator& it, octet_iterator end)
+    {
+        if (++it == end)
+            return NOT_ENOUGH_ROOM;
+
+        if (!utf8::internal::is_trail(*it))
+            return INCOMPLETE_SEQUENCE;
+
+        return UTF8_OK;
+    }
+
+    #define UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(IT, END) {utf_error ret = increase_safely(IT, END); if (ret != UTF8_OK) return ret;}    
+
+    /// get_sequence_x functions decode utf-8 sequences of the length x
+    template <typename octet_iterator>
+    utf_error get_sequence_1(octet_iterator& it, octet_iterator end, uint32_t& code_point)
+    {
+        if (it == end)
+            return NOT_ENOUGH_ROOM;
+
+        code_point = utf8::internal::mask8(*it);
+
+        return UTF8_OK;
+    }
+
+    template <typename octet_iterator>
+    utf_error get_sequence_2(octet_iterator& it, octet_iterator end, uint32_t& code_point)
+    {
+        if (it == end) 
+            return NOT_ENOUGH_ROOM;
+
+        code_point = utf8::internal::mask8(*it);
+
+        UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end)
+
+        code_point = ((code_point << 6) & 0x7ff) + ((*it) & 0x3f);
+
+        return UTF8_OK;
+    }
+
+    template <typename octet_iterator>
+    utf_error get_sequence_3(octet_iterator& it, octet_iterator end, uint32_t& code_point)
+    {
+        if (it == end)
+            return NOT_ENOUGH_ROOM;
+            
+        code_point = utf8::internal::mask8(*it);
+
+        UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end)
+
+        code_point = ((code_point << 12) & 0xffff) + ((utf8::internal::mask8(*it) << 6) & 0xfff);
+
+        UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end)
+
+        code_point += (*it) & 0x3f;
+
+        return UTF8_OK;
+    }
+
+    template <typename octet_iterator>
+    utf_error get_sequence_4(octet_iterator& it, octet_iterator end, uint32_t& code_point)
+    {
+        if (it == end)
+           return NOT_ENOUGH_ROOM;
+
+        code_point = utf8::internal::mask8(*it);
+
+        UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end)
+
+        code_point = ((code_point << 18) & 0x1fffff) + ((utf8::internal::mask8(*it) << 12) & 0x3ffff);
+
+        UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end)
+
+        code_point += (utf8::internal::mask8(*it) << 6) & 0xfff;
+
+        UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR(it, end)
+
+        code_point += (*it) & 0x3f;
+
+        return UTF8_OK;
+    }
+
+    #undef UTF8_CPP_INCREASE_AND_RETURN_ON_ERROR
+
+    template <typename octet_iterator>
+    utf_error validate_next(octet_iterator& it, octet_iterator end, uint32_t& code_point)
+    {
+        if (it == end)
+            return NOT_ENOUGH_ROOM;
+
+        // Save the original value of it so we can go back in case of failure
+        // Of course, it does not make much sense with i.e. stream iterators
+        octet_iterator original_it = it;
+
+        uint32_t cp = 0;
+        // Determine the sequence length based on the lead octet
+        typedef typename std::iterator_traits<octet_iterator>::difference_type octet_difference_type;
+        const octet_difference_type length = utf8::internal::sequence_length(it);
+
+        // Get trail octets and calculate the code point
+        utf_error err = UTF8_OK;
+        switch (length) {
+            case 0:
+                return INVALID_LEAD;
+            case 1:
+                err = utf8::internal::get_sequence_1(it, end, cp);
+                break;
+            case 2:
+                err = utf8::internal::get_sequence_2(it, end, cp);
+            break;
+            case 3:
+                err = utf8::internal::get_sequence_3(it, end, cp);
+            break;
+            case 4:
+                err = utf8::internal::get_sequence_4(it, end, cp);
+            break;
+        }
+
+        if (err == UTF8_OK) {
+            // Decoding succeeded. Now, security checks...
+            if (utf8::internal::is_code_point_valid(cp)) {
+                if (!utf8::internal::is_overlong_sequence(cp, length)){
+                    // Passed! Return here.
+                    code_point = cp;
+                    ++it;
+                    return UTF8_OK;
+                }
+                else
+                    err = OVERLONG_SEQUENCE;
+            }
+            else 
+                err = INVALID_CODE_POINT;
+        }
+
+        // Failure branch - restore the original value of the iterator
+        it = original_it;
+        return err;
+    }
+
+    template <typename octet_iterator>
+    inline utf_error validate_next(octet_iterator& it, octet_iterator end) {
+        uint32_t ignored;
+        return utf8::internal::validate_next(it, end, ignored);
+    }
+
+} // namespace internal
+}	
+
+#endif // header guard
+

third_party/utfcpp/source/utf8/unchecked.h

@@ -0,0 +1,90 @@
+// Copyright 2006 Nemanja Trifunovic
+
+/*
+Permission is hereby granted, free of charge, to any person or organization
+obtaining a copy of the software and accompanying documentation covered by
+this license (the "Software") to use, reproduce, display, distribute,
+execute, and transmit the Software, and to prepare derivative works of the
+Software, and to permit third-parties to whom the Software is furnished to
+do so, all subject to the following:
+
+The copyright notices in the Software and this entire statement, including
+the above license grant, this restriction and the following disclaimer,
+must be included in all copies of the Software, in whole or in part, and
+all derivative works of the Software, unless such copies or derivative
+works are solely in the form of machine-executable object code generated by
+a source language processor.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE, TITLE AND NON-INFRINGEMENT. IN NO EVENT
+SHALL THE COPYRIGHT HOLDERS OR ANYONE DISTRIBUTING THE SOFTWARE BE LIABLE
+FOR ANY DAMAGES OR OTHER LIABILITY, WHETHER IN CONTRACT, TORT OR OTHERWISE,
+ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+DEALINGS IN THE SOFTWARE.
+*/
+
+
+#ifndef UTF8_FOR_CPP_UNCHECKED_H_2675DCD0_9480_4c0c_B92A_CC14C027B731
+#define UTF8_FOR_CPP_UNCHECKED_H_2675DCD0_9480_4c0c_B92A_CC14C027B731
+
+#include "core.h"
+
+namespace utf8
+{
+    namespace unchecked
+    {
+        template <typename octet_iterator>
+        uint32_t next(octet_iterator& it)
+        {
+            uint32_t cp = utf8::internal::mask8(*it);
+            typename std::iterator_traits<octet_iterator>::difference_type length = utf8::internal::sequence_length(it);
+            switch (length) {
+                case 1:
+                    break;
+                case 2:
+                    it++;
+                    cp = ((cp << 6) & 0x7ff) + ((*it) & 0x3f);
+                    break;
+                case 3:
+                    ++it; 
+                    cp = ((cp << 12) & 0xffff) + ((utf8::internal::mask8(*it) << 6) & 0xfff);
+                    ++it;
+                    cp += (*it) & 0x3f;
+                    break;
+                case 4:
+                    ++it;
+                    cp = ((cp << 18) & 0x1fffff) + ((utf8::internal::mask8(*it) << 12) & 0x3ffff);                
+                    ++it;
+                    cp += (utf8::internal::mask8(*it) << 6) & 0xfff;
+                    ++it;
+                    cp += (*it) & 0x3f; 
+                    break;
+            }
+            ++it;
+            return cp;
+        }
+
+        template <typename octet_iterator>
+        uint32_t prior(octet_iterator& it)
+        {
+            while (utf8::internal::is_trail(*(--it))) ;
+            octet_iterator temp = it;
+            return utf8::unchecked::next(temp);
+        }
+
+        template <typename octet_iterator>
+        typename std::iterator_traits<octet_iterator>::difference_type
+        distance (octet_iterator first, octet_iterator last)
+        {
+            typename std::iterator_traits<octet_iterator>::difference_type dist;
+            for (dist = 0; first < last; ++dist) 
+                utf8::unchecked::next(first);
+            return dist;
+        }
+    } // namespace utf8::unchecked
+} // namespace utf8 
+
+
+#endif // header guard
+