Added osgText::String implementation to better handle decoding of encoded text.

This commit is contained in:
Robert Osfield 2003-03-05 21:05:37 +00:00
parent f9d8f3fa9d
commit 63df52d408
7 changed files with 433 additions and 70 deletions

View File

@ -3,6 +3,16 @@
OSG News (most significant items from ChangeLog)
================================================
Improved thread safety when working multipipe systems.
New OpenProducer based examples.
New MD2 plugin
New osgText implementions.
24th January 2003 - OpenSceneGraph-0.9.3.tar.gz
>>> adds support navigational light points, multi-threaded dynamic

View File

@ -105,6 +105,10 @@ SOURCE=..\..\src\osgText\Text.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\osgText\String.cpp
# End Source File
# Begin Source File
SOURCE=..\..\src\osgText\Version.cpp
# End Source File
# End Group
@ -121,6 +125,10 @@ SOURCE=..\..\include\osgText\Font
# End Source File
# Begin Source File
SOURCE=..\..\include\osgText\String
# End Source File
# Begin Source File
SOURCE=..\..\include\osgText\Text
# End Source File
# Begin Source File

75
include/osgText/String Normal file
View File

@ -0,0 +1,75 @@
/* -*-c++-*- OpenSceneGraph - Copyright (C) 1998-2003 Robert Osfield
*
* This library is open source and may be redistributed and/or modified under
* the terms of the OpenSceneGraph Public License (OSGPL) version 0.0 or
* (at your option) any later version. The full license is in LICENSE file
* included with this distribution, and on the openscenegraph.org website.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* OpenSceneGraph Public License for more details.
*/
#ifndef OSGTEXT_STRING
#define OSGTEXT_STRING 1
#include <osg/Referenced>
#include <osgText/Export>
#include <vector>
#include <set>
#include <string>
namespace osgText {
class Text;
class OSGTEXT_EXPORT String : public osg::Referenced, public std::vector<unsigned int>
{
public:
String() {}
String(const String& str);
virtual ~String() {} // public temporily while osgText is still in flux.
String& operator = (const String& str);
void set(const std::string& str);
/** Set the text using a wchar_t string,
* which is converted to an internal TextString.*/
void set(const wchar_t* text);
/**
* Types of string encodings supported
*/
enum Encoding
{
ENCODING_UNDEFINED, /// not using Unicode
ENCODING_ASCII = ENCODING_UNDEFINED,/// unsigned char ASCII
ENCODING_UTF8, /// 8-bit unicode transformation format
ENCODING_UTF16, /// 16-bit signature
ENCODING_UTF16_BE, /// 16-bit big-endian
ENCODING_UTF16_LE, /// 16-bit little-endian
ENCODING_UTF32, /// 32-bit signature
ENCODING_UTF32_BE, /// 32-bit big-endian
ENCODING_UTF32_LE, /// 32-bit little-endian
ENCODING_SIGNATURE /// detect encoding from signature
};
/** Set the text using a Unicode encoded std::string, which is converted to an internal TextString.
* The encoding parameter specificies which Unicode encodeding is used in the std::string. */
void set(const std::string& text,Encoding encoding);
/** returns a UTF8 encoded version of this osgText::String.*/
std::string createUTF8EncodedString() const;
protected:
};
}
#endif

View File

@ -18,6 +18,7 @@
#include <osg/Quat>
#include <osgText/Font>
#include <osgText/String>
namespace osgText {
@ -56,44 +57,23 @@ public:
unsigned int getFontHeight() const { return _fontWidth; }
/** TextString is a general purpose vector of char codes (unsigned int's)
* which is used internally by Text to represent strings.*/
typedef std::vector<unsigned int> TextString;
/** Set the text using a TextString.*/
void setText(const TextString& text);
/** Set the text using a osgText::String.*/
void setText(const String& text);
/** Set the text using a std::string,
* which is converted to an internal TextString.*/
void setText(const std::string& text);
/**
* Types of string encodings supported
*/
enum Encoding
{
ENCODING_UNDEFINED, /// not using Unicode
ENCODING_ASCII = ENCODING_UNDEFINED,/// unsigned char ASCII
ENCODING_UTF8, /// 8-bit unicode transformation format
ENCODING_UTF16, /// 16-bit signature
ENCODING_UTF16_BE, /// 16-bit big-endian
ENCODING_UTF16_LE, /// 16-bit little-endian
ENCODING_UTF32, /// 32-bit signature
ENCODING_UTF32_BE, /// 32-bit big-endian
ENCODING_UTF32_LE, /// 32-bit little-endian
ENCODING_SIGNATURE /// detect encoding from signature
};
/** Set the text using a Unicode encoded std::string, which is converted to an internal TextString.
* The encoding parameter specificies which Unicode encodeding is used in the std::string. */
void setText(const std::string& text,Encoding encoding);
void setText(const std::string& text,String::Encoding encoding);
/** Set the text using a wchar_t string,
* which is converted to an internal TextString.*/
void setText(const wchar_t* text);
/** Get the const text string.*/
const TextString& getText() const { return _text; }
const String& getText() const { return _text; }
/** Set the rendered character size in object coordinates.*/
@ -197,20 +177,20 @@ protected:
// members which have public access.
osg::ref_ptr<Font> _font;
unsigned int _fontWidth;
unsigned int _fontHeight;
float _characterHeight;
float _characterAspectRatio;
osg::ref_ptr<Font> _font;
unsigned int _fontWidth;
unsigned int _fontHeight;
float _characterHeight;
float _characterAspectRatio;
TextString _text;
osg::Vec3 _position;
AlignmentType _alignment;
AxisAlignment _axisAlignment;
osg::Quat _rotation;
Layout _layout;
osg::Vec4 _color;
unsigned int _drawMode;
String _text;
osg::Vec3 _position;
AlignmentType _alignment;
AxisAlignment _axisAlignment;
osg::Quat _rotation;
Layout _layout;
osg::Vec4 _color;
unsigned int _drawMode;
// internal structures, variable and methods used for rendering of characters.
struct GlyphQuads

View File

@ -6,6 +6,7 @@ CXXFILES = \
DefaultFont.cpp\
Font.cpp\
Text.cpp\
String.cpp\
Version.cpp\
DEF += -DOSGTEXT_LIBRARY

314
src/osgText/String.cpp Normal file
View File

@ -0,0 +1,314 @@
#include <osgText/String>
#include <osg/Math>
#include <osg/Notify>
using namespace osgText;
////////////////////////////////////////////////////////////////////////
//
// helper class to make it safer to querry std::string's for encoding.
//
struct look_ahead_iterator
{
look_ahead_iterator(const std::string& string):
_string(string),
_index(0),
_nullCharacter(0) {}
bool valid() const { return _index<_string.length(); }
look_ahead_iterator& operator ++ ()
{
if (_index<_string.length()) ++_index;
return *this;
}
look_ahead_iterator operator ++ (int)
{
look_ahead_iterator tmp(*this);
if (_index<_string.length()) ++_index;
return tmp;
}
look_ahead_iterator& operator += (int offset)
{
if (_index<_string.length()) _index = osg::minimum(_index+offset,_string.length());
return *this;
}
unsigned char operator * () const
{
if (_index<_string.length()) return _string[_index];
else return _nullCharacter;
}
unsigned char operator [] (unsigned int offset) const
{
if (_index+offset<_string.length()) return _string[_index+offset];
else return _nullCharacter;
}
const std::string& _string;
unsigned int _index;
unsigned char _nullCharacter;
};
String::Encoding findEncoding(look_ahead_iterator& charString,String::Encoding overrideEncoding)
{
switch (charString[0])
{
case 0xEF: // 8-bit encoding
{
// 8-bit signature = EF BB BF
if ((charString[1]==0xBB) && (charString[2]==0xBF))
{
charString+=3;
return String::ENCODING_UTF8;
}
break;
}
case 0xFE: // big-endian 16-bit
{
// 16-bit signature = FE FF
if (charString[1]==0xFF)
{
charString+=2;
return String::ENCODING_UTF16_BE;
}
break;
}
case 0xFF: // little-endian
{
// 16-bit signature = FF FE
// 32-bit signature = FF FE 00 00
if (charString[1]==0xFE)
{
// NOTE: There is an a potential problem as a 16-bit empty string
// is identical to a 32-bit start signature
if ((charString[2]==0) && (charString[3]==0) && (overrideEncoding != String::ENCODING_UTF16)) //32-bit
{
charString+=4;
return String::ENCODING_UTF32_LE;
}
else //16-bit
{
charString+=2;
return String::ENCODING_UTF16_LE;
}
}
break;
}
case 0x00: // 32-bit big-endian
{
// 32-bit signature = 00 00 FE FF
if ((charString[1]==0x00) && (charString[2]==0xFE) && (charString[3]==0xFF))
{
charString+=4;
return String::ENCODING_UTF32_BE;
}
break;
}
}
return String::ENCODING_ASCII;
}
unsigned int getNextCharacter(look_ahead_iterator& charString,String::Encoding encoding)
{
// For more info on unicode encodings see:
// http://www-106.ibm.com/developerworks/unicode/library/u-encode.html
switch(encoding)
{
case String::ENCODING_ASCII:
{
return *charString++;
}
case String::ENCODING_UTF8:
{
int char0 = *charString++;
if (char0 < 0x80) // 1-byte character
{
return char0;
}
int char1 = *charString++;
if (char0<0xe0) // 2-byte character
{
return ((char0&0x1f)<<6) | (char1&0x3f);
}
int char2 = *charString++;
if (char0<0xf0) // 3-byte character
{
return ((char0&0xf)<<12) | ((char1&0x3f)<<6) | (char2&0x3f);
}
int char3 = *charString++;
if (char0<0xf8) // 4-byte character
{
return ((char0&0x7)<<18) | ((char1&0x3f)<<12) | ((char2&0x3f)<<6) | (char3&0x3f);
}
break;
}
case String::ENCODING_UTF16_BE:
{
int char0 = *charString++;
int char1 = *charString++;
if ((char0<=0xD7) || (char0>=0xE0)) // simple character
{
return (char0<<8) | char1;
}
else if ((char0>=0xD8)&&(char0<=0xDB)) //using planes (this should get called very rarely)
{
int char2 = *charString++;
int char3 = *charString++;
int highSurrogate = (char0<<8) | char1;
int lowSurrogate = (char2<<8) | char3;
if ((char2>=0xDC)&&(char2<=0xDF)) //only for the valid range of low surrogate
{
// This covers the range of all 17 unicode planes
return ((highSurrogate-0xD800)*0x400) + (lowSurrogate-0xD800) + 0x10000;
}
}
break;
}
case String::ENCODING_UTF16_LE:
{
int char1 = *charString++;
int char0 = *charString++;
if ((char0<=0xD7) || (char0>=0xE0)) // simple character
{
return (char0<<8) | char1;
}
else if ((char0>=0xD8)&&(char0<=0xDB)) //using planes (this should get called very rarely)
{
int char3 = *charString++;
int char2 = *charString++;
int highSurrogate = (char0<<8) | char1;
int lowSurrogate = (char2<<8) | char3;
if ((char2>=0xDC)&&(char2<=0xDF)) //only for the valid range of low surrogate
{
// This covers the range of all 17 unicode planes
return ((highSurrogate-0xD800)*0x400) + (lowSurrogate-0xD800) + 0x10000;
}
}
break;
}
case String::ENCODING_UTF32_BE:
{
int character = ((((int)charString[0])<<24) | (((int)charString[1])<<16) |
(((int)charString[2])<<8) | charString[3]);
charString+=4;
if (character<0x110000)
{
// Character is constrained to the range set by the unicode standard
return character;
}
break;
}
case String::ENCODING_UTF32_LE:
{
int character = ((((int)charString[3])<<24) | (((int)charString[2])<<16) |
(((int)charString[1])<<8) | charString[0]);
charString+=4;
if (character<0x110000)
{
// Character is constrained to the range set by the unicode standard
return character;
}
break;
}
default:
{
// Should not reach this point unless the encoding is unhandled
// ENCODING_UTF16, ENCODING_UTF32 and ENCODING_SIGNATURE should never enter this method
osg::notify(osg::FATAL)<<"Error: Invalid string encoding"<<std::endl;
break;
}
}
return 0;
}
////////////////////////////////////////////////////////////////////////////
//
// String implemention.
//
String::String(const String& str):
Referenced(),
std::vector<unsigned int>(str)
{
}
String& String::operator = (const String& str)
{
if (&str==this) return *this;
clear();
std::copy(str.begin(),str.end(),std::back_inserter(*this));
return *this;
}
void String::set(const std::string& text)
{
clear();
std::copy(text.begin(),text.end(),std::back_inserter(*this));
}
void String::set(const wchar_t* text)
{
clear();
while(*text)
{
push_back(*text++);
}
}
void String::set(const std::string& text,Encoding encoding)
{
clear();
look_ahead_iterator itr(text);
if ((encoding == ENCODING_SIGNATURE) ||
(encoding == ENCODING_UTF16) ||
(encoding == ENCODING_UTF32))
{
encoding = findEncoding(itr,encoding);
}
while(itr.valid())
{
unsigned int c = getNextCharacter(itr,encoding);
if (c) push_back(c);
}
}
std::string String::createUTF8EncodedString() const
{
std::string utf8string;
for(const_iterator itr=begin();
itr!=end();
++itr)
{
unsigned int currentChar = *itr;
if (currentChar < 0x80)
{
utf8string+=(char)currentChar;
}
else if (currentChar < 0x800)
{
utf8string+=(char)(0xc0 | (currentChar>>6));
utf8string+=(char)(0x80 | currentChar & 0x3f);
}
else
{
utf8string+=(char)(0xe0 | (currentChar>>12));
utf8string+=(char)(0x80 | (currentChar>>6) & 0x3f);
utf8string+=(char)(0x80 | currentChar & 0x3f);
}
}
return utf8string;
}

View File

@ -95,7 +95,7 @@ void Text::setCharacterSize(float height,float aspectRatio)
}
void Text::setText(const TextString& text)
void Text::setText(const String& text)
{
_text = text;
computeGlyphRepresentation();
@ -103,45 +103,20 @@ void Text::setText(const TextString& text)
void Text::setText(const std::string& text)
{
_text.clear();
std::copy(text.begin(),text.end(),std::back_inserter(_text));
_text.set(text);
computeGlyphRepresentation();
}
void Text::setText(const std::string& text,Encoding encoding)
void Text::setText(const std::string& text,String::Encoding encoding)
{
_text.clear();
if (text.empty()) return;
std::cerr << "Text::setText(const std::string& text,Encoding encoding) not implemented yet."<<std::endl;
//std::string::const_iterator itr = text.begin();
if ((encoding == ENCODING_SIGNATURE) ||
(encoding == ENCODING_UTF16) ||
(encoding == ENCODING_UTF32))
{
// encoding = findEncoding(text,pos);
}
_text.set(text,encoding);
computeGlyphRepresentation();
}
void Text::setText(const wchar_t* text)
{
_text.clear();
if (text)
{
// find the end of wchar_t string
const wchar_t* endOfText = text;
while (*endOfText) ++endOfText;
// pass it to the _text field.
std::copy(text,endOfText,std::back_inserter(_text));
}
_text.set(text);
computeGlyphRepresentation();
}
@ -249,7 +224,7 @@ void Text::computeGlyphRepresentation()
float hr = _characterHeight/(float)activefont->getHeight();
float wr = hr/_characterAspectRatio;
for(TextString::iterator itr=_text.begin();
for(String::iterator itr=_text.begin();
itr!=_text.end();
++itr)
{