**********************************************************

** Big Maps Projects code (incl. Multiplayer v1.5 **
**********************************************************
- Merged Big Maps Project code from BMP+MP trunk (Revision: 3340)
o Complete SVN Revision history: https://81.169.133.124/source/ja2/branches/Wanne/JA2%201.13%20MP
- Before THIS merge, I made a branch of the existing 1.13 source
o SVN Branch: https://81.169.133.124/source/ja2/branches/JA2_rev.3336/src
- Removed old VS 6.0 and VS 2003 project and solutions files, because compilation is broken long time ago
- I will add VS 2010 projects and solution file in the next few days

git-svn-id: https://ja2svn.mooo.com/source/ja2/trunk/GameSource/ja2_v1.13/Build@3341 3b4a5df2-a311-0410-b5c6-a8a6f20db521

ext/utf8/source/utf8/checked.h

@@ -64,7 +64,7 @@ namespace utf8
     };
 
     /// The library API - functions intended to be called by the users
- 
+
     template <typename octet_iterator, typename output_iterator>
     output_iterator replace_invalid(octet_iterator start, octet_iterator end, output_iterator out, uint32_t replacement)
     {
@@ -72,7 +72,7 @@ namespace utf8
             octet_iterator sequence_start = start;
             internal::utf_error err_code = internal::validate_next(start, end);
             switch (err_code) {
-                case internal::OK :
+                case internal::UTF8_OK :
                     for (octet_iterator it = sequence_start; it != start; ++it)
                         *out++ = *it;
                     break;
@@ -92,7 +92,7 @@ namespace utf8
                         ++start;
                     break;
             }
-        }   
+        }
         return out;
     }
 
@@ -106,24 +106,24 @@ namespace utf8
     template <typename octet_iterator>
     octet_iterator append(uint32_t cp, octet_iterator result)
     {
-        if (!internal::is_code_point_valid(cp)) 
+        if (!internal::is_code_point_valid(cp))
             throw invalid_code_point(cp);
 
         if (cp < 0x80)                        // one octet
-            *(result++) = static_cast<uint8_t>(cp);  
+            *(result++) = static_cast<uint8_t>(cp);
         else if (cp < 0x800) {                // two octets
             *(result++) = static_cast<uint8_t>((cp >> 6)            | 0xc0);
             *(result++) = static_cast<uint8_t>((cp & 0x3f)          | 0x80);
         }
         else if (cp < 0x10000) {              // three octets
             *(result++) = static_cast<uint8_t>((cp >> 12)           | 0xe0);
-            *(result++) = static_cast<uint8_t>((cp >> 6) & 0x3f     | 0x80);
+            *(result++) = static_cast<uint8_t>(((cp >> 6) & 0x3f)   | 0x80);
             *(result++) = static_cast<uint8_t>((cp & 0x3f)          | 0x80);
         }
         else if (cp <= internal::CODE_POINT_MAX) {      // four octets
             *(result++) = static_cast<uint8_t>((cp >> 18)           | 0xf0);
-            *(result++) = static_cast<uint8_t>((cp >> 12)& 0x3f     | 0x80);
-            *(result++) = static_cast<uint8_t>((cp >> 6) & 0x3f     | 0x80);
+            *(result++) = static_cast<uint8_t>(((cp >> 12) & 0x3f)  | 0x80);
+            *(result++) = static_cast<uint8_t>(((cp >> 6) & 0x3f)   | 0x80);
             *(result++) = static_cast<uint8_t>((cp & 0x3f)          | 0x80);
         }
         else
@@ -138,7 +138,7 @@ namespace utf8
         uint32_t cp = 0;
         internal::utf_error err_code = internal::validate_next(it, end, &cp);
         switch (err_code) {
-            case internal::OK :
+            case internal::UTF8_OK :
                 break;
             case internal::NOT_ENOUGH_ROOM :
                 throw not_enough_room();
@@ -149,7 +149,7 @@ namespace utf8
             case internal::INVALID_CODE_POINT :
                 throw invalid_code_point(cp);
         }
-        return cp;        
+        return cp;
     }
 
     template <typename octet_iterator>
@@ -162,7 +162,7 @@ namespace utf8
     uint32_t prior(octet_iterator& it, octet_iterator start)
     {
         octet_iterator end = it;
-        while (internal::is_trail(*(--it))) 
+        while (internal::is_trail(*(--it)))
             if (it < start)
                 throw invalid_utf8(*it); // error - no lead byte in the sequence
         octet_iterator temp = it;
@@ -174,7 +174,7 @@ namespace utf8
     uint32_t previous(octet_iterator& it, octet_iterator pass_start)
     {
         octet_iterator end = it;
-        while (internal::is_trail(*(--it))) 
+        while (internal::is_trail(*(--it)))
             if (it == pass_start)
                 throw invalid_utf8(*it); // error - no lead byte in the sequence
         octet_iterator temp = it;
@@ -193,14 +193,14 @@ namespace utf8
     distance (octet_iterator first, octet_iterator last)
     {
         typename std::iterator_traits<octet_iterator>::difference_type dist;
-        for (dist = 0; first < last; ++dist) 
+        for (dist = 0; first < last; ++dist)
             next(first, last);
         return dist;
     }
 
     template <typename u16bit_iterator, typename octet_iterator>
     octet_iterator utf16to8 (u16bit_iterator start, u16bit_iterator end, octet_iterator result)
-    {       
+    {
         while (start != end) {
             uint32_t cp = internal::mask16(*start++);
             // Take care of surrogate pairs first
@@ -208,17 +208,17 @@ namespace utf8
                 if (start != end) {
                     uint32_t trail_surrogate = internal::mask16(*start++);
                     if (trail_surrogate >= internal::TRAIL_SURROGATE_MIN && trail_surrogate <= internal::TRAIL_SURROGATE_MAX)
-                        cp = (cp << 10) + trail_surrogate + internal::SURROGATE_OFFSET;                    
-                    else 
+                        cp = (cp << 10) + trail_surrogate + internal::SURROGATE_OFFSET;
+                    else
                         throw invalid_utf16(static_cast<uint16_t>(trail_surrogate));
                 }
-                else 
+                else
                     throw invalid_utf16(static_cast<uint16_t>(*start));
-            
+
             }
             result = append(cp, result);
         }
-        return result;        
+        return result;
     }
 
     template <typename u16bit_iterator, typename octet_iterator>
@@ -256,13 +256,13 @@ namespace utf8
 
     // The iterator class
     template <typename octet_iterator>
-    class iterator : public std::iterator <std::bidirectional_iterator_tag, uint32_t> { 
+    class iterator : public std::iterator <std::bidirectional_iterator_tag, uint32_t> {
       octet_iterator it;
       octet_iterator range_start;
       octet_iterator range_end;
       public:
       iterator () {};
-      explicit iterator (const octet_iterator& octet_it, 
+      explicit iterator (const octet_iterator& octet_it,
                          const octet_iterator& range_start,
                          const octet_iterator& range_end) :
                it(octet_it), range_start(range_start), range_end(range_end)
@@ -277,8 +277,8 @@ namespace utf8
           octet_iterator temp = it;
           return next(temp, range_end);
       }
-      bool operator == (const iterator& rhs) const 
-      { 
+      bool operator == (const iterator& rhs) const
+      {
           if (range_start != rhs.range_start || range_end != rhs.range_end)
               throw std::logic_error("Comparing utf-8 iterators defined with different ranges");
           return (it == rhs.it);
@@ -287,7 +287,7 @@ namespace utf8
       {
           return !(operator == (rhs));
       }
-      iterator& operator ++ () 
+      iterator& operator ++ ()
       {
           next(it, range_end);
           return *this;
@@ -297,7 +297,7 @@ namespace utf8
           iterator temp = *this;
           next(it, range_end);
           return temp;
-      }  
+      }
       iterator& operator -- ()
       {
           prior(it, range_start);

ext/utf8/source/utf8/core.h

@@ -33,7 +33,7 @@ DEALINGS IN THE SOFTWARE.
 namespace utf8
 {
     // The typedefs for 8-bit, 16-bit and 32-bit unsigned integers
-    // You may need to change them to match your system. 
+    // 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;
@@ -41,7 +41,7 @@ namespace utf8
 
 // 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
@@ -81,14 +81,14 @@ namespace internal
     inline bool is_code_point_valid(u32 cp)
     {
         return (cp <= CODE_POINT_MAX && !is_surrogate(cp) && cp != 0xfffe && cp != 0xffff);
-    }  
+    }
 
     template <typename octet_iterator>
     inline typename std::iterator_traits<octet_iterator>::difference_type
     sequence_length(octet_iterator lead_it)
     {
         uint8_t lead = mask8(*lead_it);
-        if (lead < 0x80) 
+        if (lead < 0x80)
             return 1;
         else if ((lead >> 5) == 0x6)
             return 2;
@@ -96,121 +96,196 @@ namespace internal
             return 3;
         else if ((lead >> 3) == 0x1e)
             return 4;
-        else 
+        else
             return 0;
     }
 
-    enum utf_error {OK, NOT_ENOUGH_ROOM, INVALID_LEAD, INCOMPLETE_SEQUENCE, OVERLONG_SEQUENCE, INVALID_CODE_POINT};
+    inline bool is_overlong_sequence(uint32_t cp, int 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};
+
+    /// 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) {
+            if (code_point)
+                *code_point = mask8(*it);
+            return UTF8_OK;
+        }
+        return NOT_ENOUGH_ROOM;
+    }
+
+    template <typename octet_iterator>
+    utf_error get_sequence_2(octet_iterator& it, octet_iterator end, uint32_t* code_point)
+    {
+        utf_error ret_code = NOT_ENOUGH_ROOM;
+
+        if (it != end) {
+            uint32_t cp = mask8(*it);
+            if (++it != end) {
+                if (is_trail(*it)) {
+                    cp = ((cp << 6) & 0x7ff) + ((*it) & 0x3f);
+
+                    if (code_point)
+                        *code_point = cp;
+                    ret_code = UTF8_OK;
+                }
+                else
+                    ret_code = INCOMPLETE_SEQUENCE;
+            }
+            else
+                ret_code = NOT_ENOUGH_ROOM;
+        }
+
+        return ret_code;
+    }
+
+    template <typename octet_iterator>
+    utf_error get_sequence_3(octet_iterator& it, octet_iterator end, uint32_t* code_point)
+    {
+        utf_error ret_code = NOT_ENOUGH_ROOM;
+
+        if (it != end) {
+            uint32_t cp = mask8(*it);
+            if (++it != end) {
+                if (is_trail(*it)) {
+                    cp = ((cp << 12) & 0xffff) + ((mask8(*it) << 6) & 0xfff);
+                    if (++it != end) {
+                        if (is_trail(*it)) {
+                            cp += (*it) & 0x3f;
+
+                            if (code_point)
+                                *code_point = cp;
+                            ret_code = UTF8_OK;
+                        }
+                        else 
+                            ret_code = INCOMPLETE_SEQUENCE;
+                    }
+                    else
+                        ret_code = NOT_ENOUGH_ROOM;
+                }
+                else
+                    ret_code = INCOMPLETE_SEQUENCE;
+            }
+            else
+                ret_code = NOT_ENOUGH_ROOM;
+        }
+
+        return ret_code;
+    }
+
+    template <typename octet_iterator>
+    utf_error get_sequence_4(octet_iterator& it, octet_iterator end, uint32_t* code_point)
+    {
+        utf_error ret_code = NOT_ENOUGH_ROOM;
+
+        if (it != end) {
+            uint32_t cp = mask8(*it);
+            if (++it != end) {
+                if (is_trail(*it)) {
+                    cp = ((cp << 18) & 0x1fffff) + ((mask8(*it) << 12) & 0x3ffff);
+                    if (++it != end) {
+                        if (is_trail(*it)) {
+                            cp += (mask8(*it) << 6) & 0xfff;
+                            if (++it != end) {
+                                if (is_trail(*it)) {
+                                    cp += (*it) & 0x3f;
+
+                                    if (code_point)
+                                        *code_point = cp;
+                                    ret_code = UTF8_OK;
+                                }
+                                else
+                                    ret_code = INCOMPLETE_SEQUENCE;
+                            }
+                            else
+                                ret_code = NOT_ENOUGH_ROOM;
+                        }
+                        else
+                            ret_code = INCOMPLETE_SEQUENCE;
+                    }
+                    else
+                        ret_code = NOT_ENOUGH_ROOM;
+                }
+                else 
+                    ret_code = INCOMPLETE_SEQUENCE;
+            }
+            else
+                ret_code = NOT_ENOUGH_ROOM;
+        }
+
+        return ret_code;
+    }
 
     template <typename octet_iterator>
     utf_error validate_next(octet_iterator& it, octet_iterator end, uint32_t* code_point)
     {
-        uint32_t cp = mask8(*it);
-        // Check the lead octet
+        // 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;
         octet_difference_type length = sequence_length(it);
+        if (length == 0)
+            return INVALID_LEAD;
 
-        // "Shortcut" for ASCII characters
-        if (length == 1) {
-            if (end - it > 0) {
-                if (code_point)
-                    *code_point = cp;
-                ++it;
-                return OK;
-            }
-            else
-                return NOT_ENOUGH_ROOM;
-        }
-
-        // Do we have enough memory?     
-        if (std::distance(it, end) < length)
-            return NOT_ENOUGH_ROOM;
-        
-        // Check trail octets and calculate the code point
+        // Now that we have a valid sequence length, get trail octets and calculate the code point
+        utf_error err = UTF8_OK;
         switch (length) {
-            case 0:
-                return INVALID_LEAD;
+            case 1:
+                err = get_sequence_1(it, end, &cp);
                 break;
             case 2:
-                if (is_trail(*(++it))) { 
-                    cp = ((cp << 6) & 0x7ff) + ((*it) & 0x3f);
-                }
-                else {
-                    --it;
-                    return INCOMPLETE_SEQUENCE;
-                }
+                err = get_sequence_2(it, end, &cp);
             break;
             case 3:
-                if (is_trail(*(++it))) {
-                    cp = ((cp << 12) & 0xffff) + ((mask8(*it) << 6) & 0xfff);
-                    if (is_trail(*(++it))) {
-                        cp += (*it) & 0x3f;
-                    }
-                    else {
-                        std::advance(it, -2);
-                        return INCOMPLETE_SEQUENCE;
-                    }
-                }
-                else {
-                    --it;
-                    return INCOMPLETE_SEQUENCE;
-                }
+                err = get_sequence_3(it, end, &cp);
             break;
             case 4:
-                if (is_trail(*(++it))) {
-                    cp = ((cp << 18) & 0x1fffff) + ((mask8(*it) << 12) & 0x3ffff);                
-                    if (is_trail(*(++it))) {
-                        cp += (mask8(*it) << 6) & 0xfff;
-                        if (is_trail(*(++it))) {
-                            cp += (*it) & 0x3f; 
-                        }
-                        else {
-                            std::advance(it, -3);
-                            return INCOMPLETE_SEQUENCE;
-                        }
-                    }
-                    else {
-                        std::advance(it, -2);
-                        return INCOMPLETE_SEQUENCE;
-                    }
-                }
-                else {
-                    --it;
-                    return INCOMPLETE_SEQUENCE;
-                }
+                err = get_sequence_4(it, end, &cp);
             break;
         }
-        // Is the code point valid?
-        if (!is_code_point_valid(cp)) {
-            for (octet_difference_type i = 0; i < length - 1; ++i) 
-                --it;
-            return INVALID_CODE_POINT;
-        }
-            
-        if (code_point)
-            *code_point = cp;
-            
-        if (cp < 0x80) {
-            if (length != 1) {
-                std::advance(it, -(length-1));
-                return OVERLONG_SEQUENCE;
+
+        if (err == UTF8_OK) {
+            // Decoding succeeded. Now, security checks...
+            if (is_code_point_valid(cp)) {
+                if (!is_overlong_sequence(cp, length)){
+                    // Passed! Return here.
+                    if (code_point)
+                        *code_point = cp;
+                    ++it;
+                    return UTF8_OK;
+                }
+                else
+                    err = OVERLONG_SEQUENCE;
             }
+            else 
+                err = INVALID_CODE_POINT;
         }
-        else if (cp < 0x800) {
-            if (length != 2) {
-                std::advance(it, -(length-1));
-                return OVERLONG_SEQUENCE;
-            }
-        }
-        else if (cp < 0x10000) {
-            if (length != 3) {
-                std::advance(it, -(length-1));
-                return OVERLONG_SEQUENCE;
-            }
-        }
-           
-        ++it;
-        return OK;    
+
+        // Failure branch - restore the original value of the iterator
+        it = original_it;
+        return err;
     }
 
     template <typename octet_iterator>
@@ -218,12 +293,12 @@ namespace internal
         return validate_next(it, end, 0);
     }
 
-} // namespace internal 
+} // namespace internal
 
     /// The library API - functions intended to be called by the users
 
     // Byte order mark
-    const uint8_t bom[] = {0xef, 0xbb, 0xbf}; 
+    const uint8_t bom[] = {0xef, 0xbb, 0xbf};
 
     template <typename octet_iterator>
     octet_iterator find_invalid(octet_iterator start, octet_iterator end)
@@ -231,7 +306,7 @@ namespace internal
         octet_iterator result = start;
         while (result != end) {
             internal::utf_error err_code = internal::validate_next(result, end);
-            if (err_code != internal::OK)
+            if (err_code != internal::UTF8_OK)
                 return result;
         }
         return result;

ext/utf8/source/utf8/unchecked.h

@@ -45,13 +45,13 @@ namespace utf8
             }
             else if (cp < 0x10000) {              // three octets
                 *(result++) = static_cast<uint8_t>((cp >> 12)         | 0xe0);
-                *(result++) = static_cast<uint8_t>((cp >> 6) & 0x3f   | 0x80);
+                *(result++) = static_cast<uint8_t>(((cp >> 6) & 0x3f) | 0x80);
                 *(result++) = static_cast<uint8_t>((cp & 0x3f)        | 0x80);
             }
             else {                                // four octets
                 *(result++) = static_cast<uint8_t>((cp >> 18)         | 0xf0);
-                *(result++) = static_cast<uint8_t>((cp >> 12)& 0x3f   | 0x80);
-                *(result++) = static_cast<uint8_t>((cp >> 6) & 0x3f   | 0x80);
+                *(result++) = static_cast<uint8_t>(((cp >> 12) & 0x3f)| 0x80);
+                *(result++) = static_cast<uint8_t>(((cp >> 6) & 0x3f) | 0x80);
                 *(result++) = static_cast<uint8_t>((cp & 0x3f)        | 0x80);
             }
             return result;