tinyxmlparser.cpp

这个代码用C＋＋实现了一个小型虚拟机

/*
www.sourceforge.net/projects/tinyxml
Original code (2.0 and earlier )copyright (c) 2000-2002 Lee Thomason (www.grinninglizard.com)

This software is provided 'as-is', without any express or implied 
warranty. In no event will the authors be held liable for any 
damages arising from the use of this software.

Permission is granted to anyone to use this software for any 
purpose, including commercial applications, and to alter it and 
redistribute it freely, subject to the following restrictions:

1. The origin of this software must not be misrepresented; you must 
not claim that you wrote the original software. If you use this
software in a product, an acknowledgment in the product documentation
would be appreciated but is not required.

2. Altered source versions must be plainly marked as such, and 
must not be misrepresented as being the original software.

3. This notice may not be removed or altered from any source 
distribution.
*/
#include "../stdafx.h"
#include "tinyxml.h"
#include <ctype.h>
#include <stddef.h>

//#define DEBUG_PARSER

// Note tha "PutString" hardcodes the same list. This
// is less flexible than it appears. Changing the entries
// or order will break putstring.	
TiXmlBase::Entity TiXmlBase::entity[ NUM_ENTITY ] = 
{
	{ "&amp;",  5, '&' },
	{ "&lt;",   4, '<' },
	{ "&gt;",   4, '>' },
	{ "&quot;", 6, '\"' },
	{ "&apos;", 6, '\'' }
};

// Bunch of unicode info at:
//		http://www.unicode.org/faq/utf_bom.html
// Including the basic of this table, which determines the #bytes in the
// sequence from the lead byte. 1 placed for invalid sequences --
// although the result will be junk, pass it through as much as possible.
// Beware of the non-characters in UTF-8:	
//				ef bb bf (Microsoft "lead bytes")
//				ef bf be
//				ef bf bf 

const unsigned char TIXML_UTF_LEAD_0 = 0xefU;
const unsigned char TIXML_UTF_LEAD_1 = 0xbbU;
const unsigned char TIXML_UTF_LEAD_2 = 0xbfU;

const int TiXmlBase::utf8ByteTable[256] = 
{
	//	0	1	2	3	4	5	6	7	8	9	a	b	c	d	e	f
		1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	// 0x00
		1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	// 0x10
		1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	// 0x20
		1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	// 0x30
		1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	// 0x40
		1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	// 0x50
		1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	// 0x60
		1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	// 0x70	End of ASCII range
		1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	// 0x80 0x80 to 0xc1 invalid
		1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	// 0x90 
		1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	// 0xa0 
		1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	// 0xb0 
		1,	1,	2,	2,	2,	2,	2,	2,	2,	2,	2,	2,	2,	2,	2,	2,	// 0xc0 0xc2 to 0xdf 2 byte
		2,	2,	2,	2,	2,	2,	2,	2,	2,	2,	2,	2,	2,	2,	2,	2,	// 0xd0
		3,	3,	3,	3,	3,	3,	3,	3,	3,	3,	3,	3,	3,	3,	3,	3,	// 0xe0 0xe0 to 0xef 3 byte
		4,	4,	4,	4,	4,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1,	1	// 0xf0 0xf0 to 0xf4 4 byte, 0xf5 and higher invalid
};


void TiXmlBase::ConvertUTF32ToUTF8( unsigned long input, char* output, int* length )
{
	const unsigned long BYTE_MASK = 0xBF;
	const unsigned long BYTE_MARK = 0x80;
	const unsigned long FIRST_BYTE_MARK[7] = { 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC };

	if (input < 0x80) 
		*length = 1;
	else if ( input < 0x800 )
		*length = 2;
	else if ( input < 0x10000 )
		*length = 3;
	else if ( input < 0x200000 )
		*length = 4;
	else
		{ *length = 0; return; }	// This code won't covert this correctly anyway.

	output += *length;

	// Scary scary fall throughs.
	switch (*length) 
	{
		case 4:
			--output; 
			*output = (char)((input | BYTE_MARK) & BYTE_MASK); 
			input >>= 6;
		case 3:
			--output; 
			*output = (char)((input | BYTE_MARK) & BYTE_MASK); 
			input >>= 6;
		case 2:
			--output; 
			*output = (char)((input | BYTE_MARK) & BYTE_MASK); 
			input >>= 6;
		case 1:
			--output; 
			*output = (char)(input | FIRST_BYTE_MARK[*length]);
	}
}


/*static*/ int TiXmlBase::IsAlpha( unsigned char anyByte, TiXmlEncoding /*encoding*/ )
{
	// This will only work for low-ascii, everything else is assumed to be a valid
	// letter. I'm not sure this is the best approach, but it is quite tricky trying
	// to figure out alhabetical vs. not across encoding. So take a very 
	// conservative approach.

//	if ( encoding == TIXML_ENCODING_UTF8 )
//	{
		if ( anyByte < 127 )
			return isalpha( anyByte );
		else
			return 1;	// What else to do? The unicode set is huge...get the english ones right.
//	}
//	else
//	{
//		return isalpha( anyByte );
//	}
}


/*static*/ int TiXmlBase::IsAlphaNum( unsigned char anyByte, TiXmlEncoding /*encoding*/ )
{
	// This will only work for low-ascii, everything else is assumed to be a valid
	// letter. I'm not sure this is the best approach, but it is quite tricky trying
	// to figure out alhabetical vs. not across encoding. So take a very 
	// conservative approach.

//	if ( encoding == TIXML_ENCODING_UTF8 )
//	{
		if ( anyByte < 127 )
			return isalnum( anyByte );
		else
			return 1;	// What else to do? The unicode set is huge...get the english ones right.
//	}
//	else
//	{
//		return isalnum( anyByte );
//	}
}


class TiXmlParsingData
{
	friend class TiXmlDocument;
  public:
	void Stamp( const char* now, TiXmlEncoding encoding );

	const TiXmlCursor& Cursor()	{ return cursor; }

  private:
	// Only used by the document!
	TiXmlParsingData( const char* start, int _tabsize, int row, int col )
	{
		assert( start );
		stamp = start;
		tabsize = _tabsize;
		cursor.row = row;
		cursor.col = col;
	}

	TiXmlCursor		cursor;
	const char*		stamp;
	int				tabsize;
};


void TiXmlParsingData::Stamp( const char* now, TiXmlEncoding encoding )
{
	assert( now );

	// Do nothing if the tabsize is 0.
	if ( tabsize < 1 )
	{
		return;
	}

	// Get the current row, column.
	int row = cursor.row;
	int col = cursor.col;
	const char* p = stamp;
	assert( p );

	while ( p < now )
	{
		// Treat p as unsigned, so we have a happy compiler.
		const unsigned char* pU = (const unsigned char*)p;

		// Code contributed by Fletcher Dunn: (modified by lee)
		switch (*pU) {
			case 0:
				// We *should* never get here, but in case we do, don't
				// advance past the terminating null character, ever
				return;

			case '\r':
				// bump down to the next line
				++row;
				col = 0;				
				// Eat the character
				++p;

				// Check for \r\n sequence, and treat this as a single character
				if (*p == '\n') {
					++p;
				}
				break;

			case '\n':
				// bump down to the next line
				++row;
				col = 0;

				// Eat the character
				++p;

				// Check for \n\r sequence, and treat this as a single
				// character.  (Yes, this bizarre thing does occur still
				// on some arcane platforms...)
				if (*p == '\r') {
					++p;
				}
				break;

			case '\t':
				// Eat the character
				++p;

				// Skip to next tab stop
				col = (col / tabsize + 1) * tabsize;
				break;

			case TIXML_UTF_LEAD_0:
				if ( encoding == TIXML_ENCODING_UTF8 )
				{
					if ( *(p+1) && *(p+2) )
					{
						// In these cases, don't advance the column. These are
						// 0-width spaces.
						if ( *(pU+1)==TIXML_UTF_LEAD_1 && *(pU+2)==TIXML_UTF_LEAD_2 )
							p += 3;	
						else if ( *(pU+1)==0xbfU && *(pU+2)==0xbeU )
							p += 3;	
						else if ( *(pU+1)==0xbfU && *(pU+2)==0xbfU )
							p += 3;	
						else
							{ p +=3; ++col; }	// A normal character.
					}
				}
				else
				{
					++p;
					++col;
				}
				break;

			default:
				if ( encoding == TIXML_ENCODING_UTF8 )
				{
					// Eat the 1 to 4 byte utf8 character.
					int step = TiXmlBase::utf8ByteTable[*((unsigned char*)p)];
					if ( step == 0 )
						step = 1;		// Error case from bad encoding, but handle gracefully.
					p += step;

					// Just advance one column, of course.
					++col;
				}
				else
				{
					++p;
					++col;
				}
				break;
		}
	}
	cursor.row = row;
	cursor.col = col;
	assert( cursor.row >= -1 );
	assert( cursor.col >= -1 );
	stamp = p;
	assert( stamp );
}


const char* TiXmlBase::SkipWhiteSpace( const char* p, TiXmlEncoding encoding )
{
	if ( !p || !*p )
	{
		return 0;
	}
	if ( encoding == TIXML_ENCODING_UTF8 )
	{
		while ( *p )
		{
			const unsigned char* pU = (const unsigned char*)p;
			
			// Skip the stupid Microsoft UTF-8 Byte order marks
			if (	*(pU+0)==TIXML_UTF_LEAD_0
				 && *(pU+1)==TIXML_UTF_LEAD_1 
				 && *(pU+2)==TIXML_UTF_LEAD_2 )
			{
				p += 3;
				continue;
			}
			else if(*(pU+0)==TIXML_UTF_LEAD_0
				 && *(pU+1)==0xbfU
				 && *(pU+2)==0xbeU )
			{
				p += 3;
				continue;
			}
			else if(*(pU+0)==TIXML_UTF_LEAD_0
				 && *(pU+1)==0xbfU
				 && *(pU+2)==0xbfU )
			{
				p += 3;
				continue;
			}

			if ( IsWhiteSpace( *p ) || *p == '\n' || *p =='\r' )		// Still using old rules for white space.
				++p;
			else
				break;
		}
	}
	else
	{
		while ( *p && IsWhiteSpace( *p ) || *p == '\n' || *p =='\r' )
			++p;
	}

	return p;
}

#ifdef TIXML_USE_STL
/*static*/ bool TiXmlBase::StreamWhiteSpace( TIXML_ISTREAM * in, TIXML_STRING * tag )
{
	for( ;; )
	{
		if ( !in->good() ) return false;

		int c = in->peek();
		// At this scope, we can't get to a document. So fail silently.
		if ( !IsWhiteSpace( c ) || c <= 0 )
			return true;

		*tag += (char) in->get();
	}
}

/*static*/ bool TiXmlBase::StreamTo( TIXML_ISTREAM * in, int character, TIXML_STRING * tag )
{
	//assert( character > 0 && character < 128 );	// else it won't work in utf-8
	while ( in->good() )
	{
		int c = in->peek();
		if ( c == character )
			return true;
		if ( c <= 0 )		// Silent failure: can't get document at this scope
			return false;

		in->get();
		*tag += (char) c;
	}
	return false;
}
#endif

const char* TiXmlBase::ReadName( const char* p, TIXML_STRING * name, TiXmlEncoding encoding )
{
	*name = "";
	assert( p );

	// Names start with letters or underscores.
	// Of course, in unicode, tinyxml has no idea what a letter *is*. The
	// algorithm is generous.
	//
	// After that, they can be letters, underscores, numbers,
	// hyphens, or colons. (Colons are valid ony for namespaces,
	// but tinyxml can't tell namespaces from names.)
	if (    p && *p 
		 && ( IsAlpha( (unsigned char) *p, encoding ) || *p == '_' ) )
	{
		while(		p && *p
				&&	(		IsAlphaNum( (unsigned char ) *p, encoding ) 
						 || *p == '_'
						 || *p == '-'
						 || *p == '.'
						 || *p == ':' ) )
		{
			(*name) += *p;
			++p;
		}
		return p;
	}
	return 0;
}

const char* TiXmlBase::GetEntity( const char* p, char* value, int* length, TiXmlEncoding encoding )
{
	// Presume an entity, and pull it out.
    TIXML_STRING ent;
	int i;
	*length = 0;

	if ( *(p+1) && *(p+1) == '#' && *(p+2) )
	{
		unsigned long ucs = 0;
		ptrdiff_t delta = 0;
		unsigned mult = 1;

		if ( *(p+2) == 'x' )
		{
			// Hexadecimal.
			if ( !*(p+3) ) return 0;

			const char* q = p+3;
			q = strchr( q, ';' );

			if ( !q || !*q ) return 0;

			delta = q-p;
			--q;

			while ( *q != 'x' )
			{
				if ( *q >= '0' && *q <= '9' )
					ucs += mult * (*q - '0');
				else if ( *q >= 'a' && *q <= 'f' )
					ucs += mult * (*q - 'a' + 10);
				else if ( *q >= 'A' && *q <= 'F' )
					ucs += mult * (*q - 'A' + 10 );
				else 
					return 0;
				mult *= 16;
				--q;
			}
		}
		else
		{
			// Decimal.
			if ( !*(p+2) ) return 0;

			const char* q = p+2;
			q = strchr( q, ';' );

			if ( !q || !*q ) return 0;

			delta = q-p;
			--q;

			while ( *q != '#' )
			{
				if ( *q >= '0' && *q <= '9' )
					ucs += mult * (*q - '0');
				else 
					return 0;
				mult *= 10;
				--q;
			}
		}
		if ( encoding == TIXML_ENCODING_UTF8 )
		{
			// convert the UCS to UTF-8
			ConvertUTF32ToUTF8( ucs, value, length );
		}
		else
		{
			*value = (char)ucs;
			*length = 1;
		}
		return p + delta + 1;
	}

	// Now try to match it.
	for( i=0; i<NUM_ENTITY; ++i )
	{
		if ( strncmp( entity[i].str, p, entity[i].strLength ) == 0 )
		{
			assert( strlen( entity[i].str ) == entity[i].strLength );
			*value = entity[i].chr;
			*length = 1;
			return ( p + entity[i].strLength );
		}
	}

	// So it wasn't an entity, its unrecognized, or something like that.
	*value = *p;	// Don't put back the last one, since we return it!
	return p+1;
}


bool TiXmlBase::StringEqual( const char* p,
							 const char* tag,
							 bool ignoreCase,
							 TiXmlEncoding encoding )
{
	assert( p );
	assert( tag );
	if ( !p || !*p )
	{
		assert( 0 );
		return false;
	}