tinyxml | Mbed

Users » steiger » Code » tinyxml

tinyxmlparser.cpp@0:7c97dcef700c, 2010-11-21 (annotated)

Committer:: steiger
Date:: Sun Nov 21 00:49:17 2010 +0000
Revision:: 0:7c97dcef700c

Who changed what in which revision?

User	Revision	Line number	New contents of line
steiger	0:7c97dcef700c	1	/*
steiger	0:7c97dcef700c	2	www.sourceforge.net/projects/tinyxml
steiger	0:7c97dcef700c	3	Original code (2.0 and earlier )copyright (c) 2000-2002 Lee Thomason (www.grinninglizard.com)
steiger	0:7c97dcef700c	4
steiger	0:7c97dcef700c	5	This software is provided 'as-is', without any express or implied
steiger	0:7c97dcef700c	6	warranty. In no event will the authors be held liable for any
steiger	0:7c97dcef700c	7	damages arising from the use of this software.
steiger	0:7c97dcef700c	8
steiger	0:7c97dcef700c	9	Permission is granted to anyone to use this software for any
steiger	0:7c97dcef700c	10	purpose, including commercial applications, and to alter it and
steiger	0:7c97dcef700c	11	redistribute it freely, subject to the following restrictions:
steiger	0:7c97dcef700c	12
steiger	0:7c97dcef700c	13	1. The origin of this software must not be misrepresented; you must
steiger	0:7c97dcef700c	14	not claim that you wrote the original software. If you use this
steiger	0:7c97dcef700c	15	software in a product, an acknowledgment in the product documentation
steiger	0:7c97dcef700c	16	would be appreciated but is not required.
steiger	0:7c97dcef700c	17
steiger	0:7c97dcef700c	18	2. Altered source versions must be plainly marked as such, and
steiger	0:7c97dcef700c	19	must not be misrepresented as being the original software.
steiger	0:7c97dcef700c	20
steiger	0:7c97dcef700c	21	3. This notice may not be removed or altered from any source
steiger	0:7c97dcef700c	22	distribution.
steiger	0:7c97dcef700c	23	*/
steiger	0:7c97dcef700c	24
steiger	0:7c97dcef700c	25	#include <ctype.h>
steiger	0:7c97dcef700c	26	#include <stddef.h>
steiger	0:7c97dcef700c	27
steiger	0:7c97dcef700c	28	#include "tinyxml.h"
steiger	0:7c97dcef700c	29
steiger	0:7c97dcef700c	30	//#define DEBUG_PARSER
steiger	0:7c97dcef700c	31	#if defined( DEBUG_PARSER )
steiger	0:7c97dcef700c	32	# if defined( DEBUG ) && defined( _MSC_VER )
steiger	0:7c97dcef700c	33	# include <windows.h>
steiger	0:7c97dcef700c	34	# define TIXML_LOG OutputDebugString
steiger	0:7c97dcef700c	35	# else
steiger	0:7c97dcef700c	36	# define TIXML_LOG printf
steiger	0:7c97dcef700c	37	# endif
steiger	0:7c97dcef700c	38	#endif
steiger	0:7c97dcef700c	39
steiger	0:7c97dcef700c	40	// Note tha "PutString" hardcodes the same list. This
steiger	0:7c97dcef700c	41	// is less flexible than it appears. Changing the entries
steiger	0:7c97dcef700c	42	// or order will break putstring.
steiger	0:7c97dcef700c	43	TiXmlBase::Entity TiXmlBase::entity[ NUM_ENTITY ] =
steiger	0:7c97dcef700c	44	{
steiger	0:7c97dcef700c	45	{ "&", 5, '&' },
steiger	0:7c97dcef700c	46	{ "<", 4, '<' },
steiger	0:7c97dcef700c	47	{ ">", 4, '>' },
steiger	0:7c97dcef700c	48	{ """, 6, '\"' },
steiger	0:7c97dcef700c	49	{ "'", 6, '\'' }
steiger	0:7c97dcef700c	50	};
steiger	0:7c97dcef700c	51
steiger	0:7c97dcef700c	52	// Bunch of unicode info at:
steiger	0:7c97dcef700c	53	// http://www.unicode.org/faq/utf_bom.html
steiger	0:7c97dcef700c	54	// Including the basic of this table, which determines the #bytes in the
steiger	0:7c97dcef700c	55	// sequence from the lead byte. 1 placed for invalid sequences --
steiger	0:7c97dcef700c	56	// although the result will be junk, pass it through as much as possible.
steiger	0:7c97dcef700c	57	// Beware of the non-characters in UTF-8:
steiger	0:7c97dcef700c	58	// ef bb bf (Microsoft "lead bytes")
steiger	0:7c97dcef700c	59	// ef bf be
steiger	0:7c97dcef700c	60	// ef bf bf
steiger	0:7c97dcef700c	61
steiger	0:7c97dcef700c	62	const unsigned char TIXML_UTF_LEAD_0 = 0xefU;
steiger	0:7c97dcef700c	63	const unsigned char TIXML_UTF_LEAD_1 = 0xbbU;
steiger	0:7c97dcef700c	64	const unsigned char TIXML_UTF_LEAD_2 = 0xbfU;
steiger	0:7c97dcef700c	65
steiger	0:7c97dcef700c	66	const int TiXmlBase::utf8ByteTable[256] =
steiger	0:7c97dcef700c	67	{
steiger	0:7c97dcef700c	68	// 0 1 2 3 4 5 6 7 8 9 a b c d e f
steiger	0:7c97dcef700c	69	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, // 0x00
steiger	0:7c97dcef700c	70	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, // 0x10
steiger	0:7c97dcef700c	71	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, // 0x20
steiger	0:7c97dcef700c	72	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, // 0x30
steiger	0:7c97dcef700c	73	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, // 0x40
steiger	0:7c97dcef700c	74	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, // 0x50
steiger	0:7c97dcef700c	75	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, // 0x60
steiger	0:7c97dcef700c	76	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, // 0x70 End of ASCII range
steiger	0:7c97dcef700c	77	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, // 0x80 0x80 to 0xc1 invalid
steiger	0:7c97dcef700c	78	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, // 0x90
steiger	0:7c97dcef700c	79	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, // 0xa0
steiger	0:7c97dcef700c	80	1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, // 0xb0
steiger	0:7c97dcef700c	81	1, 1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, // 0xc0 0xc2 to 0xdf 2 byte
steiger	0:7c97dcef700c	82	2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, // 0xd0
steiger	0:7c97dcef700c	83	3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3, // 0xe0 0xe0 to 0xef 3 byte
steiger	0:7c97dcef700c	84	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
steiger	0:7c97dcef700c	85	};
steiger	0:7c97dcef700c	86
steiger	0:7c97dcef700c	87
steiger	0:7c97dcef700c	88	void TiXmlBase::ConvertUTF32ToUTF8( unsigned long input, char* output, int* length )
steiger	0:7c97dcef700c	89	{
steiger	0:7c97dcef700c	90	const unsigned long BYTE_MASK = 0xBF;
steiger	0:7c97dcef700c	91	const unsigned long BYTE_MARK = 0x80;
steiger	0:7c97dcef700c	92	const unsigned long FIRST_BYTE_MARK[7] = { 0x00, 0x00, 0xC0, 0xE0, 0xF0, 0xF8, 0xFC };
steiger	0:7c97dcef700c	93
steiger	0:7c97dcef700c	94	if (input < 0x80)
steiger	0:7c97dcef700c	95	*length = 1;
steiger	0:7c97dcef700c	96	else if ( input < 0x800 )
steiger	0:7c97dcef700c	97	*length = 2;
steiger	0:7c97dcef700c	98	else if ( input < 0x10000 )
steiger	0:7c97dcef700c	99	*length = 3;
steiger	0:7c97dcef700c	100	else if ( input < 0x200000 )
steiger	0:7c97dcef700c	101	*length = 4;
steiger	0:7c97dcef700c	102	else
steiger	0:7c97dcef700c	103	{ *length = 0; return; } // This code won't covert this correctly anyway.
steiger	0:7c97dcef700c	104
steiger	0:7c97dcef700c	105	output += *length;
steiger	0:7c97dcef700c	106
steiger	0:7c97dcef700c	107	// Scary scary fall throughs.
steiger	0:7c97dcef700c	108	switch (*length)
steiger	0:7c97dcef700c	109	{
steiger	0:7c97dcef700c	110	case 4:
steiger	0:7c97dcef700c	111	--output;
steiger	0:7c97dcef700c	112	*output = (char)((input \| BYTE_MARK) & BYTE_MASK);
steiger	0:7c97dcef700c	113	input >>= 6;
steiger	0:7c97dcef700c	114	case 3:
steiger	0:7c97dcef700c	115	--output;
steiger	0:7c97dcef700c	116	*output = (char)((input \| BYTE_MARK) & BYTE_MASK);
steiger	0:7c97dcef700c	117	input >>= 6;
steiger	0:7c97dcef700c	118	case 2:
steiger	0:7c97dcef700c	119	--output;
steiger	0:7c97dcef700c	120	*output = (char)((input \| BYTE_MARK) & BYTE_MASK);
steiger	0:7c97dcef700c	121	input >>= 6;
steiger	0:7c97dcef700c	122	case 1:
steiger	0:7c97dcef700c	123	--output;
steiger	0:7c97dcef700c	124	output = (char)(input \| FIRST_BYTE_MARK[length]);
steiger	0:7c97dcef700c	125	}
steiger	0:7c97dcef700c	126	}
steiger	0:7c97dcef700c	127
steiger	0:7c97dcef700c	128
steiger	0:7c97dcef700c	129	/static/ int TiXmlBase::IsAlpha( unsigned char anyByte, TiXmlEncoding /encoding/ )
steiger	0:7c97dcef700c	130	{
steiger	0:7c97dcef700c	131	// This will only work for low-ascii, everything else is assumed to be a valid
steiger	0:7c97dcef700c	132	// letter. I'm not sure this is the best approach, but it is quite tricky trying
steiger	0:7c97dcef700c	133	// to figure out alhabetical vs. not across encoding. So take a very
steiger	0:7c97dcef700c	134	// conservative approach.
steiger	0:7c97dcef700c	135
steiger	0:7c97dcef700c	136	// if ( encoding == TIXML_ENCODING_UTF8 )
steiger	0:7c97dcef700c	137	// {
steiger	0:7c97dcef700c	138	if ( anyByte < 127 )
steiger	0:7c97dcef700c	139	return isalpha( anyByte );
steiger	0:7c97dcef700c	140	else
steiger	0:7c97dcef700c	141	return 1; // What else to do? The unicode set is huge...get the english ones right.
steiger	0:7c97dcef700c	142	// }
steiger	0:7c97dcef700c	143	// else
steiger	0:7c97dcef700c	144	// {
steiger	0:7c97dcef700c	145	// return isalpha( anyByte );
steiger	0:7c97dcef700c	146	// }
steiger	0:7c97dcef700c	147	}
steiger	0:7c97dcef700c	148
steiger	0:7c97dcef700c	149
steiger	0:7c97dcef700c	150	/static/ int TiXmlBase::IsAlphaNum( unsigned char anyByte, TiXmlEncoding /encoding/ )
steiger	0:7c97dcef700c	151	{
steiger	0:7c97dcef700c	152	// This will only work for low-ascii, everything else is assumed to be a valid
steiger	0:7c97dcef700c	153	// letter. I'm not sure this is the best approach, but it is quite tricky trying
steiger	0:7c97dcef700c	154	// to figure out alhabetical vs. not across encoding. So take a very
steiger	0:7c97dcef700c	155	// conservative approach.
steiger	0:7c97dcef700c	156
steiger	0:7c97dcef700c	157	// if ( encoding == TIXML_ENCODING_UTF8 )
steiger	0:7c97dcef700c	158	// {
steiger	0:7c97dcef700c	159	if ( anyByte < 127 )
steiger	0:7c97dcef700c	160	return isalnum( anyByte );
steiger	0:7c97dcef700c	161	else
steiger	0:7c97dcef700c	162	return 1; // What else to do? The unicode set is huge...get the english ones right.
steiger	0:7c97dcef700c	163	// }
steiger	0:7c97dcef700c	164	// else
steiger	0:7c97dcef700c	165	// {
steiger	0:7c97dcef700c	166	// return isalnum( anyByte );
steiger	0:7c97dcef700c	167	// }
steiger	0:7c97dcef700c	168	}
steiger	0:7c97dcef700c	169
steiger	0:7c97dcef700c	170
steiger	0:7c97dcef700c	171	class TiXmlParsingData
steiger	0:7c97dcef700c	172	{
steiger	0:7c97dcef700c	173	friend class TiXmlDocument;
steiger	0:7c97dcef700c	174	public:
steiger	0:7c97dcef700c	175	void Stamp( const char* now, TiXmlEncoding encoding );
steiger	0:7c97dcef700c	176
steiger	0:7c97dcef700c	177	const TiXmlCursor& Cursor() { return cursor; }
steiger	0:7c97dcef700c	178
steiger	0:7c97dcef700c	179	private:
steiger	0:7c97dcef700c	180	// Only used by the document!
steiger	0:7c97dcef700c	181	TiXmlParsingData( const char* start, int _tabsize, int row, int col )
steiger	0:7c97dcef700c	182	{
steiger	0:7c97dcef700c	183	assert( start );
steiger	0:7c97dcef700c	184	stamp = start;
steiger	0:7c97dcef700c	185	tabsize = _tabsize;
steiger	0:7c97dcef700c	186	cursor.row = row;
steiger	0:7c97dcef700c	187	cursor.col = col;
steiger	0:7c97dcef700c	188	}
steiger	0:7c97dcef700c	189
steiger	0:7c97dcef700c	190	TiXmlCursor cursor;
steiger	0:7c97dcef700c	191	const char* stamp;
steiger	0:7c97dcef700c	192	int tabsize;
steiger	0:7c97dcef700c	193	};
steiger	0:7c97dcef700c	194
steiger	0:7c97dcef700c	195
steiger	0:7c97dcef700c	196	void TiXmlParsingData::Stamp( const char* now, TiXmlEncoding encoding )
steiger	0:7c97dcef700c	197	{
steiger	0:7c97dcef700c	198	assert( now );
steiger	0:7c97dcef700c	199
steiger	0:7c97dcef700c	200	// Do nothing if the tabsize is 0.
steiger	0:7c97dcef700c	201	if ( tabsize < 1 )
steiger	0:7c97dcef700c	202	{
steiger	0:7c97dcef700c	203	return;
steiger	0:7c97dcef700c	204	}
steiger	0:7c97dcef700c	205
steiger	0:7c97dcef700c	206	// Get the current row, column.
steiger	0:7c97dcef700c	207	int row = cursor.row;
steiger	0:7c97dcef700c	208	int col = cursor.col;
steiger	0:7c97dcef700c	209	const char* p = stamp;
steiger	0:7c97dcef700c	210	assert( p );
steiger	0:7c97dcef700c	211
steiger	0:7c97dcef700c	212	while ( p < now )
steiger	0:7c97dcef700c	213	{
steiger	0:7c97dcef700c	214	// Treat p as unsigned, so we have a happy compiler.
steiger	0:7c97dcef700c	215	const unsigned char* pU = (const unsigned char*)p;
steiger	0:7c97dcef700c	216
steiger	0:7c97dcef700c	217	// Code contributed by Fletcher Dunn: (modified by lee)
steiger	0:7c97dcef700c	218	switch (*pU) {
steiger	0:7c97dcef700c	219	case 0:
steiger	0:7c97dcef700c	220	// We should never get here, but in case we do, don't
steiger	0:7c97dcef700c	221	// advance past the terminating null character, ever
steiger	0:7c97dcef700c	222	return;
steiger	0:7c97dcef700c	223
steiger	0:7c97dcef700c	224	case '\r':
steiger	0:7c97dcef700c	225	// bump down to the next line
steiger	0:7c97dcef700c	226	++row;
steiger	0:7c97dcef700c	227	col = 0;
steiger	0:7c97dcef700c	228	// Eat the character
steiger	0:7c97dcef700c	229	++p;
steiger	0:7c97dcef700c	230
steiger	0:7c97dcef700c	231	// Check for \r\n sequence, and treat this as a single character
steiger	0:7c97dcef700c	232	if (*p == '\n') {
steiger	0:7c97dcef700c	233	++p;
steiger	0:7c97dcef700c	234	}
steiger	0:7c97dcef700c	235	break;
steiger	0:7c97dcef700c	236
steiger	0:7c97dcef700c	237	case '\n':
steiger	0:7c97dcef700c	238	// bump down to the next line
steiger	0:7c97dcef700c	239	++row;
steiger	0:7c97dcef700c	240	col = 0;
steiger	0:7c97dcef700c	241
steiger	0:7c97dcef700c	242	// Eat the character
steiger	0:7c97dcef700c	243	++p;
steiger	0:7c97dcef700c	244
steiger	0:7c97dcef700c	245	// Check for \n\r sequence, and treat this as a single
steiger	0:7c97dcef700c	246	// character. (Yes, this bizarre thing does occur still
steiger	0:7c97dcef700c	247	// on some arcane platforms...)
steiger	0:7c97dcef700c	248	if (*p == '\r') {
steiger	0:7c97dcef700c	249	++p;
steiger	0:7c97dcef700c	250	}
steiger	0:7c97dcef700c	251	break;
steiger	0:7c97dcef700c	252
steiger	0:7c97dcef700c	253	case '\t':
steiger	0:7c97dcef700c	254	// Eat the character
steiger	0:7c97dcef700c	255	++p;
steiger	0:7c97dcef700c	256
steiger	0:7c97dcef700c	257	// Skip to next tab stop
steiger	0:7c97dcef700c	258	col = (col / tabsize + 1) * tabsize;
steiger	0:7c97dcef700c	259	break;
steiger	0:7c97dcef700c	260
steiger	0:7c97dcef700c	261	case TIXML_UTF_LEAD_0:
steiger	0:7c97dcef700c	262	if ( encoding == TIXML_ENCODING_UTF8 )
steiger	0:7c97dcef700c	263	{
steiger	0:7c97dcef700c	264	if ( (p+1) && (p+2) )
steiger	0:7c97dcef700c	265	{
steiger	0:7c97dcef700c	266	// In these cases, don't advance the column. These are
steiger	0:7c97dcef700c	267	// 0-width spaces.
steiger	0:7c97dcef700c	268	if ( (pU+1)==TIXML_UTF_LEAD_1 && (pU+2)==TIXML_UTF_LEAD_2 )
steiger	0:7c97dcef700c	269	p += 3;
steiger	0:7c97dcef700c	270	else if ( (pU+1)==0xbfU && (pU+2)==0xbeU )
steiger	0:7c97dcef700c	271	p += 3;
steiger	0:7c97dcef700c	272	else if ( (pU+1)==0xbfU && (pU+2)==0xbfU )
steiger	0:7c97dcef700c	273	p += 3;
steiger	0:7c97dcef700c	274	else
steiger	0:7c97dcef700c	275	{ p +=3; ++col; } // A normal character.
steiger	0:7c97dcef700c	276	}
steiger	0:7c97dcef700c	277	}
steiger	0:7c97dcef700c	278	else
steiger	0:7c97dcef700c	279	{
steiger	0:7c97dcef700c	280	++p;
steiger	0:7c97dcef700c	281	++col;
steiger	0:7c97dcef700c	282	}
steiger	0:7c97dcef700c	283	break;
steiger	0:7c97dcef700c	284
steiger	0:7c97dcef700c	285	default:
steiger	0:7c97dcef700c	286	if ( encoding == TIXML_ENCODING_UTF8 )
steiger	0:7c97dcef700c	287	{
steiger	0:7c97dcef700c	288	// Eat the 1 to 4 byte utf8 character.
steiger	0:7c97dcef700c	289	int step = TiXmlBase::utf8ByteTable[((const unsigned char)p)];
steiger	0:7c97dcef700c	290	if ( step == 0 )
steiger	0:7c97dcef700c	291	step = 1; // Error case from bad encoding, but handle gracefully.
steiger	0:7c97dcef700c	292	p += step;
steiger	0:7c97dcef700c	293
steiger	0:7c97dcef700c	294	// Just advance one column, of course.
steiger	0:7c97dcef700c	295	++col;
steiger	0:7c97dcef700c	296	}
steiger	0:7c97dcef700c	297	else
steiger	0:7c97dcef700c	298	{
steiger	0:7c97dcef700c	299	++p;
steiger	0:7c97dcef700c	300	++col;
steiger	0:7c97dcef700c	301	}
steiger	0:7c97dcef700c	302	break;
steiger	0:7c97dcef700c	303	}
steiger	0:7c97dcef700c	304	}
steiger	0:7c97dcef700c	305	cursor.row = row;
steiger	0:7c97dcef700c	306	cursor.col = col;
steiger	0:7c97dcef700c	307	assert( cursor.row >= -1 );
steiger	0:7c97dcef700c	308	assert( cursor.col >= -1 );
steiger	0:7c97dcef700c	309	stamp = p;
steiger	0:7c97dcef700c	310	assert( stamp );
steiger	0:7c97dcef700c	311	}
steiger	0:7c97dcef700c	312
steiger	0:7c97dcef700c	313
steiger	0:7c97dcef700c	314	const char* TiXmlBase::SkipWhiteSpace( const char* p, TiXmlEncoding encoding )
steiger	0:7c97dcef700c	315	{
steiger	0:7c97dcef700c	316	if ( !p \|\| !*p )
steiger	0:7c97dcef700c	317	{
steiger	0:7c97dcef700c	318	return 0;
steiger	0:7c97dcef700c	319	}
steiger	0:7c97dcef700c	320	if ( encoding == TIXML_ENCODING_UTF8 )
steiger	0:7c97dcef700c	321	{
steiger	0:7c97dcef700c	322	while ( *p )
steiger	0:7c97dcef700c	323	{
steiger	0:7c97dcef700c	324	const unsigned char* pU = (const unsigned char*)p;
steiger	0:7c97dcef700c	325
steiger	0:7c97dcef700c	326	// Skip the stupid Microsoft UTF-8 Byte order marks
steiger	0:7c97dcef700c	327	if ( *(pU+0)==TIXML_UTF_LEAD_0
steiger	0:7c97dcef700c	328	&& *(pU+1)==TIXML_UTF_LEAD_1
steiger	0:7c97dcef700c	329	&& *(pU+2)==TIXML_UTF_LEAD_2 )
steiger	0:7c97dcef700c	330	{
steiger	0:7c97dcef700c	331	p += 3;
steiger	0:7c97dcef700c	332	continue;
steiger	0:7c97dcef700c	333	}
steiger	0:7c97dcef700c	334	else if(*(pU+0)==TIXML_UTF_LEAD_0
steiger	0:7c97dcef700c	335	&& *(pU+1)==0xbfU
steiger	0:7c97dcef700c	336	&& *(pU+2)==0xbeU )
steiger	0:7c97dcef700c	337	{
steiger	0:7c97dcef700c	338	p += 3;
steiger	0:7c97dcef700c	339	continue;
steiger	0:7c97dcef700c	340	}
steiger	0:7c97dcef700c	341	else if(*(pU+0)==TIXML_UTF_LEAD_0
steiger	0:7c97dcef700c	342	&& *(pU+1)==0xbfU
steiger	0:7c97dcef700c	343	&& *(pU+2)==0xbfU )
steiger	0:7c97dcef700c	344	{
steiger	0:7c97dcef700c	345	p += 3;
steiger	0:7c97dcef700c	346	continue;
steiger	0:7c97dcef700c	347	}
steiger	0:7c97dcef700c	348
steiger	0:7c97dcef700c	349	if ( IsWhiteSpace( *p ) ) // Still using old rules for white space.
steiger	0:7c97dcef700c	350	++p;
steiger	0:7c97dcef700c	351	else
steiger	0:7c97dcef700c	352	break;
steiger	0:7c97dcef700c	353	}
steiger	0:7c97dcef700c	354	}
steiger	0:7c97dcef700c	355	else
steiger	0:7c97dcef700c	356	{
steiger	0:7c97dcef700c	357	while ( p && IsWhiteSpace( p ) )
steiger	0:7c97dcef700c	358	++p;
steiger	0:7c97dcef700c	359	}
steiger	0:7c97dcef700c	360
steiger	0:7c97dcef700c	361	return p;
steiger	0:7c97dcef700c	362	}
steiger	0:7c97dcef700c	363
steiger	0:7c97dcef700c	364	#ifdef TIXML_USE_STL
steiger	0:7c97dcef700c	365	/static/ bool TiXmlBase::StreamWhiteSpace( std::istream * in, TIXML_STRING * tag )
steiger	0:7c97dcef700c	366	{
steiger	0:7c97dcef700c	367	for( ;; )
steiger	0:7c97dcef700c	368	{
steiger	0:7c97dcef700c	369	if ( !in->good() ) return false;
steiger	0:7c97dcef700c	370
steiger	0:7c97dcef700c	371	int c = in->peek();
steiger	0:7c97dcef700c	372	// At this scope, we can't get to a document. So fail silently.
steiger	0:7c97dcef700c	373	if ( !IsWhiteSpace( c ) \|\| c <= 0 )
steiger	0:7c97dcef700c	374	return true;
steiger	0:7c97dcef700c	375
steiger	0:7c97dcef700c	376	*tag += (char) in->get();
steiger	0:7c97dcef700c	377	}
steiger	0:7c97dcef700c	378	}
steiger	0:7c97dcef700c	379
steiger	0:7c97dcef700c	380	/static/ bool TiXmlBase::StreamTo( std::istream * in, int character, TIXML_STRING * tag )
steiger	0:7c97dcef700c	381	{
steiger	0:7c97dcef700c	382	//assert( character > 0 && character < 128 ); // else it won't work in utf-8
steiger	0:7c97dcef700c	383	while ( in->good() )
steiger	0:7c97dcef700c	384	{
steiger	0:7c97dcef700c	385	int c = in->peek();
steiger	0:7c97dcef700c	386	if ( c == character )
steiger	0:7c97dcef700c	387	return true;
steiger	0:7c97dcef700c	388	if ( c <= 0 ) // Silent failure: can't get document at this scope
steiger	0:7c97dcef700c	389	return false;
steiger	0:7c97dcef700c	390
steiger	0:7c97dcef700c	391	in->get();
steiger	0:7c97dcef700c	392	*tag += (char) c;
steiger	0:7c97dcef700c	393	}
steiger	0:7c97dcef700c	394	return false;
steiger	0:7c97dcef700c	395	}
steiger	0:7c97dcef700c	396	#endif
steiger	0:7c97dcef700c	397
steiger	0:7c97dcef700c	398	// One of TinyXML's more performance demanding functions. Try to keep the memory overhead down. The
steiger	0:7c97dcef700c	399	// "assign" optimization removes over 10% of the execution time.
steiger	0:7c97dcef700c	400	//
steiger	0:7c97dcef700c	401	const char* TiXmlBase::ReadName( const char* p, TIXML_STRING * name, TiXmlEncoding encoding )
steiger	0:7c97dcef700c	402	{
steiger	0:7c97dcef700c	403	// Oddly, not supported on some comilers,
steiger	0:7c97dcef700c	404	//name->clear();
steiger	0:7c97dcef700c	405	// So use this:
steiger	0:7c97dcef700c	406	*name = "";
steiger	0:7c97dcef700c	407	assert( p );
steiger	0:7c97dcef700c	408
steiger	0:7c97dcef700c	409	// Names start with letters or underscores.
steiger	0:7c97dcef700c	410	// Of course, in unicode, tinyxml has no idea what a letter is. The
steiger	0:7c97dcef700c	411	// algorithm is generous.
steiger	0:7c97dcef700c	412	//
steiger	0:7c97dcef700c	413	// After that, they can be letters, underscores, numbers,
steiger	0:7c97dcef700c	414	// hyphens, or colons. (Colons are valid ony for namespaces,
steiger	0:7c97dcef700c	415	// but tinyxml can't tell namespaces from names.)
steiger	0:7c97dcef700c	416	if ( p && *p
steiger	0:7c97dcef700c	417	&& ( IsAlpha( (unsigned char) p, encoding ) \|\| p == '_' ) )
steiger	0:7c97dcef700c	418	{
steiger	0:7c97dcef700c	419	const char* start = p;
steiger	0:7c97dcef700c	420	while( p && *p
steiger	0:7c97dcef700c	421	&& ( IsAlphaNum( (unsigned char ) *p, encoding )
steiger	0:7c97dcef700c	422	\|\| *p == '_'
steiger	0:7c97dcef700c	423	\|\| *p == '-'
steiger	0:7c97dcef700c	424	\|\| *p == '.'
steiger	0:7c97dcef700c	425	\|\| *p == ':' ) )
steiger	0:7c97dcef700c	426	{
steiger	0:7c97dcef700c	427	//(name) += p; // expensive
steiger	0:7c97dcef700c	428	++p;
steiger	0:7c97dcef700c	429	}
steiger	0:7c97dcef700c	430	if ( p-start > 0 ) {
steiger	0:7c97dcef700c	431	name->assign( start, p-start );
steiger	0:7c97dcef700c	432	}
steiger	0:7c97dcef700c	433	return p;
steiger	0:7c97dcef700c	434	}
steiger	0:7c97dcef700c	435	return 0;
steiger	0:7c97dcef700c	436	}
steiger	0:7c97dcef700c	437
steiger	0:7c97dcef700c	438	const char* TiXmlBase::GetEntity( const char* p, char* value, int* length, TiXmlEncoding encoding )
steiger	0:7c97dcef700c	439	{
steiger	0:7c97dcef700c	440	// Presume an entity, and pull it out.
steiger	0:7c97dcef700c	441	TIXML_STRING ent;
steiger	0:7c97dcef700c	442	int i;
steiger	0:7c97dcef700c	443	*length = 0;
steiger	0:7c97dcef700c	444
steiger	0:7c97dcef700c	445	if ( (p+1) && (p+1) == '#' && *(p+2) )
steiger	0:7c97dcef700c	446	{
steiger	0:7c97dcef700c	447	unsigned long ucs = 0;
steiger	0:7c97dcef700c	448	ptrdiff_t delta = 0;
steiger	0:7c97dcef700c	449	unsigned mult = 1;
steiger	0:7c97dcef700c	450
steiger	0:7c97dcef700c	451	if ( *(p+2) == 'x' )
steiger	0:7c97dcef700c	452	{
steiger	0:7c97dcef700c	453	// Hexadecimal.
steiger	0:7c97dcef700c	454	if ( !*(p+3) ) return 0;
steiger	0:7c97dcef700c	455
steiger	0:7c97dcef700c	456	const char* q = p+3;
steiger	0:7c97dcef700c	457	q = strchr( q, ';' );
steiger	0:7c97dcef700c	458
steiger	0:7c97dcef700c	459	if ( !q \|\| !*q ) return 0;
steiger	0:7c97dcef700c	460
steiger	0:7c97dcef700c	461	delta = q-p;
steiger	0:7c97dcef700c	462	--q;
steiger	0:7c97dcef700c	463
steiger	0:7c97dcef700c	464	while ( *q != 'x' )
steiger	0:7c97dcef700c	465	{
steiger	0:7c97dcef700c	466	if ( q >= '0' && q <= '9' )
steiger	0:7c97dcef700c	467	ucs += mult * (*q - '0');
steiger	0:7c97dcef700c	468	else if ( q >= 'a' && q <= 'f' )
steiger	0:7c97dcef700c	469	ucs += mult * (*q - 'a' + 10);
steiger	0:7c97dcef700c	470	else if ( q >= 'A' && q <= 'F' )
steiger	0:7c97dcef700c	471	ucs += mult * (*q - 'A' + 10 );
steiger	0:7c97dcef700c	472	else
steiger	0:7c97dcef700c	473	return 0;
steiger	0:7c97dcef700c	474	mult *= 16;
steiger	0:7c97dcef700c	475	--q;
steiger	0:7c97dcef700c	476	}
steiger	0:7c97dcef700c	477	}
steiger	0:7c97dcef700c	478	else
steiger	0:7c97dcef700c	479	{
steiger	0:7c97dcef700c	480	// Decimal.
steiger	0:7c97dcef700c	481	if ( !*(p+2) ) return 0;
steiger	0:7c97dcef700c	482
steiger	0:7c97dcef700c	483	const char* q = p+2;
steiger	0:7c97dcef700c	484	q = strchr( q, ';' );
steiger	0:7c97dcef700c	485
steiger	0:7c97dcef700c	486	if ( !q \|\| !*q ) return 0;
steiger	0:7c97dcef700c	487
steiger	0:7c97dcef700c	488	delta = q-p;
steiger	0:7c97dcef700c	489	--q;
steiger	0:7c97dcef700c	490
steiger	0:7c97dcef700c	491	while ( *q != '#' )
steiger	0:7c97dcef700c	492	{
steiger	0:7c97dcef700c	493	if ( q >= '0' && q <= '9' )
steiger	0:7c97dcef700c	494	ucs += mult * (*q - '0');
steiger	0:7c97dcef700c	495	else
steiger	0:7c97dcef700c	496	return 0;
steiger	0:7c97dcef700c	497	mult *= 10;
steiger	0:7c97dcef700c	498	--q;
steiger	0:7c97dcef700c	499	}
steiger	0:7c97dcef700c	500	}
steiger	0:7c97dcef700c	501	if ( encoding == TIXML_ENCODING_UTF8 )
steiger	0:7c97dcef700c	502	{
steiger	0:7c97dcef700c	503	// convert the UCS to UTF-8
steiger	0:7c97dcef700c	504	ConvertUTF32ToUTF8( ucs, value, length );
steiger	0:7c97dcef700c	505	}
steiger	0:7c97dcef700c	506	else
steiger	0:7c97dcef700c	507	{
steiger	0:7c97dcef700c	508	*value = (char)ucs;
steiger	0:7c97dcef700c	509	*length = 1;
steiger	0:7c97dcef700c	510	}
steiger	0:7c97dcef700c	511	return p + delta + 1;
steiger	0:7c97dcef700c	512	}
steiger	0:7c97dcef700c	513
steiger	0:7c97dcef700c	514	// Now try to match it.
steiger	0:7c97dcef700c	515	for( i=0; i<NUM_ENTITY; ++i )
steiger	0:7c97dcef700c	516	{
steiger	0:7c97dcef700c	517	if ( strncmp( entity[i].str, p, entity[i].strLength ) == 0 )
steiger	0:7c97dcef700c	518	{
steiger	0:7c97dcef700c	519	assert( strlen( entity[i].str ) == entity[i].strLength );
steiger	0:7c97dcef700c	520	*value = entity[i].chr;
steiger	0:7c97dcef700c	521	*length = 1;
steiger	0:7c97dcef700c	522	return ( p + entity[i].strLength );
steiger	0:7c97dcef700c	523	}
steiger	0:7c97dcef700c	524	}
steiger	0:7c97dcef700c	525
steiger	0:7c97dcef700c	526	// So it wasn't an entity, its unrecognized, or something like that.
steiger	0:7c97dcef700c	527	value = p; // Don't put back the last one, since we return it!
steiger	0:7c97dcef700c	528	//*length = 1; // Leave unrecognized entities - this doesn't really work.
steiger	0:7c97dcef700c	529	// Just writes strange XML.
steiger	0:7c97dcef700c	530	return p+1;
steiger	0:7c97dcef700c	531	}
steiger	0:7c97dcef700c	532
steiger	0:7c97dcef700c	533
steiger	0:7c97dcef700c	534	bool TiXmlBase::StringEqual( const char* p,
steiger	0:7c97dcef700c	535	const char* tag,
steiger	0:7c97dcef700c	536	bool ignoreCase,
steiger	0:7c97dcef700c	537	TiXmlEncoding encoding )
steiger	0:7c97dcef700c	538	{
steiger	0:7c97dcef700c	539	assert( p );
steiger	0:7c97dcef700c	540	assert( tag );
steiger	0:7c97dcef700c	541	if ( !p \|\| !*p )
steiger	0:7c97dcef700c	542	{
steiger	0:7c97dcef700c	543	assert( 0 );
steiger	0:7c97dcef700c	544	return false;
steiger	0:7c97dcef700c	545	}
steiger	0:7c97dcef700c	546
steiger	0:7c97dcef700c	547	const char* q = p;
steiger	0:7c97dcef700c	548
steiger	0:7c97dcef700c	549	if ( ignoreCase )
steiger	0:7c97dcef700c	550	{
steiger	0:7c97dcef700c	551	while ( q && tag && ToLower( q, encoding ) == ToLower( tag, encoding ) )
steiger	0:7c97dcef700c	552	{
steiger	0:7c97dcef700c	553	++q;
steiger	0:7c97dcef700c	554	++tag;
steiger	0:7c97dcef700c	555	}
steiger	0:7c97dcef700c	556
steiger	0:7c97dcef700c	557	if ( *tag == 0 )
steiger	0:7c97dcef700c	558	return true;
steiger	0:7c97dcef700c	559	}
steiger	0:7c97dcef700c	560	else
steiger	0:7c97dcef700c	561	{
steiger	0:7c97dcef700c	562	while ( q && tag && q == tag )
steiger	0:7c97dcef700c	563	{
steiger	0:7c97dcef700c	564	++q;
steiger	0:7c97dcef700c	565	++tag;
steiger	0:7c97dcef700c	566	}
steiger	0:7c97dcef700c	567
steiger	0:7c97dcef700c	568	if ( *tag == 0 ) // Have we found the end of the tag, and everything equal?
steiger	0:7c97dcef700c	569	return true;
steiger	0:7c97dcef700c	570	}
steiger	0:7c97dcef700c	571	return false;
steiger	0:7c97dcef700c	572	}
steiger	0:7c97dcef700c	573
steiger	0:7c97dcef700c	574	const char* TiXmlBase::ReadText( const char* p,
steiger	0:7c97dcef700c	575	TIXML_STRING * text,
steiger	0:7c97dcef700c	576	bool trimWhiteSpace,
steiger	0:7c97dcef700c	577	const char* endTag,
steiger	0:7c97dcef700c	578	bool caseInsensitive,
steiger	0:7c97dcef700c	579	TiXmlEncoding encoding )
steiger	0:7c97dcef700c	580	{
steiger	0:7c97dcef700c	581	*text = "";
steiger	0:7c97dcef700c	582	if ( !trimWhiteSpace // certain tags always keep whitespace
steiger	0:7c97dcef700c	583	\|\| !condenseWhiteSpace ) // if true, whitespace is always kept
steiger	0:7c97dcef700c	584	{
steiger	0:7c97dcef700c	585	// Keep all the white space.
steiger	0:7c97dcef700c	586	while ( p && *p
steiger	0:7c97dcef700c	587	&& !StringEqual( p, endTag, caseInsensitive, encoding )
steiger	0:7c97dcef700c	588	)
steiger	0:7c97dcef700c	589	{
steiger	0:7c97dcef700c	590	int len;
steiger	0:7c97dcef700c	591	char cArr[4] = { 0, 0, 0, 0 };
steiger	0:7c97dcef700c	592	p = GetChar( p, cArr, &len, encoding );
steiger	0:7c97dcef700c	593	text->append( cArr, len );
steiger	0:7c97dcef700c	594	}
steiger	0:7c97dcef700c	595	}
steiger	0:7c97dcef700c	596	else
steiger	0:7c97dcef700c	597	{
steiger	0:7c97dcef700c	598	bool whitespace = false;
steiger	0:7c97dcef700c	599
steiger	0:7c97dcef700c	600	// Remove leading white space:
steiger	0:7c97dcef700c	601	p = SkipWhiteSpace( p, encoding );
steiger	0:7c97dcef700c	602	while ( p && *p
steiger	0:7c97dcef700c	603	&& !StringEqual( p, endTag, caseInsensitive, encoding ) )
steiger	0:7c97dcef700c	604	{
steiger	0:7c97dcef700c	605	if ( p == '\r' \|\| p == '\n' )
steiger	0:7c97dcef700c	606	{
steiger	0:7c97dcef700c	607	whitespace = true;
steiger	0:7c97dcef700c	608	++p;
steiger	0:7c97dcef700c	609	}
steiger	0:7c97dcef700c	610	else if ( IsWhiteSpace( *p ) )
steiger	0:7c97dcef700c	611	{
steiger	0:7c97dcef700c	612	whitespace = true;
steiger	0:7c97dcef700c	613	++p;
steiger	0:7c97dcef700c	614	}
steiger	0:7c97dcef700c	615	else
steiger	0:7c97dcef700c	616	{
steiger	0:7c97dcef700c	617	// If we've found whitespace, add it before the
steiger	0:7c97dcef700c	618	// new character. Any whitespace just becomes a space.
steiger	0:7c97dcef700c	619	if ( whitespace )
steiger	0:7c97dcef700c	620	{
steiger	0:7c97dcef700c	621	(*text) += ' ';
steiger	0:7c97dcef700c	622	whitespace = false;
steiger	0:7c97dcef700c	623	}
steiger	0:7c97dcef700c	624	int len;
steiger	0:7c97dcef700c	625	char cArr[4] = { 0, 0, 0, 0 };
steiger	0:7c97dcef700c	626	p = GetChar( p, cArr, &len, encoding );
steiger	0:7c97dcef700c	627	if ( len == 1 )
steiger	0:7c97dcef700c	628	(*text) += cArr[0]; // more efficient
steiger	0:7c97dcef700c	629	else
steiger	0:7c97dcef700c	630	text->append( cArr, len );
steiger	0:7c97dcef700c	631	}
steiger	0:7c97dcef700c	632	}
steiger	0:7c97dcef700c	633	}
steiger	0:7c97dcef700c	634	if ( p && *p )
steiger	0:7c97dcef700c	635	p += strlen( endTag );
steiger	0:7c97dcef700c	636	return p;
steiger	0:7c97dcef700c	637	}
steiger	0:7c97dcef700c	638
steiger	0:7c97dcef700c	639	#ifdef TIXML_USE_STL
steiger	0:7c97dcef700c	640
steiger	0:7c97dcef700c	641	void TiXmlDocument::StreamIn( std::istream * in, TIXML_STRING * tag )
steiger	0:7c97dcef700c	642	{
steiger	0:7c97dcef700c	643	// The basic issue with a document is that we don't know what we're
steiger	0:7c97dcef700c	644	// streaming. Read something presumed to be a tag (and hope), then
steiger	0:7c97dcef700c	645	// identify it, and call the appropriate stream method on the tag.
steiger	0:7c97dcef700c	646	//
steiger	0:7c97dcef700c	647	// This "pre-streaming" will never read the closing ">" so the
steiger	0:7c97dcef700c	648	// sub-tag can orient itself.
steiger	0:7c97dcef700c	649
steiger	0:7c97dcef700c	650	if ( !StreamTo( in, '<', tag ) )
steiger	0:7c97dcef700c	651	{
steiger	0:7c97dcef700c	652	SetError( TIXML_ERROR_PARSING_EMPTY, 0, 0, TIXML_ENCODING_UNKNOWN );
steiger	0:7c97dcef700c	653	return;
steiger	0:7c97dcef700c	654	}
steiger	0:7c97dcef700c	655
steiger	0:7c97dcef700c	656	while ( in->good() )
steiger	0:7c97dcef700c	657	{
steiger	0:7c97dcef700c	658	int tagIndex = (int) tag->length();
steiger	0:7c97dcef700c	659	while ( in->good() && in->peek() != '>' )
steiger	0:7c97dcef700c	660	{
steiger	0:7c97dcef700c	661	int c = in->get();
steiger	0:7c97dcef700c	662	if ( c <= 0 )
steiger	0:7c97dcef700c	663	{
steiger	0:7c97dcef700c	664	SetError( TIXML_ERROR_EMBEDDED_NULL, 0, 0, TIXML_ENCODING_UNKNOWN );
steiger	0:7c97dcef700c	665	break;
steiger	0:7c97dcef700c	666	}
steiger	0:7c97dcef700c	667	(*tag) += (char) c;
steiger	0:7c97dcef700c	668	}
steiger	0:7c97dcef700c	669
steiger	0:7c97dcef700c	670	if ( in->good() )
steiger	0:7c97dcef700c	671	{
steiger	0:7c97dcef700c	672	// We now have something we presume to be a node of
steiger	0:7c97dcef700c	673	// some sort. Identify it, and call the node to
steiger	0:7c97dcef700c	674	// continue streaming.
steiger	0:7c97dcef700c	675	TiXmlNode* node = Identify( tag->c_str() + tagIndex, TIXML_DEFAULT_ENCODING );
steiger	0:7c97dcef700c	676
steiger	0:7c97dcef700c	677	if ( node )
steiger	0:7c97dcef700c	678	{
steiger	0:7c97dcef700c	679	node->StreamIn( in, tag );
steiger	0:7c97dcef700c	680	bool isElement = node->ToElement() != 0;
steiger	0:7c97dcef700c	681	delete node;
steiger	0:7c97dcef700c	682	node = 0;
steiger	0:7c97dcef700c	683
steiger	0:7c97dcef700c	684	// If this is the root element, we're done. Parsing will be
steiger	0:7c97dcef700c	685	// done by the >> operator.
steiger	0:7c97dcef700c	686	if ( isElement )
steiger	0:7c97dcef700c	687	{
steiger	0:7c97dcef700c	688	return;
steiger	0:7c97dcef700c	689	}
steiger	0:7c97dcef700c	690	}
steiger	0:7c97dcef700c	691	else
steiger	0:7c97dcef700c	692	{
steiger	0:7c97dcef700c	693	SetError( TIXML_ERROR, 0, 0, TIXML_ENCODING_UNKNOWN );
steiger	0:7c97dcef700c	694	return;
steiger	0:7c97dcef700c	695	}
steiger	0:7c97dcef700c	696	}
steiger	0:7c97dcef700c	697	}
steiger	0:7c97dcef700c	698	// We should have returned sooner.
steiger	0:7c97dcef700c	699	SetError( TIXML_ERROR, 0, 0, TIXML_ENCODING_UNKNOWN );
steiger	0:7c97dcef700c	700	}
steiger	0:7c97dcef700c	701
steiger	0:7c97dcef700c	702	#endif
steiger	0:7c97dcef700c	703
steiger	0:7c97dcef700c	704	const char* TiXmlDocument::Parse( const char* p, TiXmlParsingData* prevData, TiXmlEncoding encoding )
steiger	0:7c97dcef700c	705	{
steiger	0:7c97dcef700c	706	ClearError();
steiger	0:7c97dcef700c	707
steiger	0:7c97dcef700c	708	// Parse away, at the document level. Since a document
steiger	0:7c97dcef700c	709	// contains nothing but other tags, most of what happens
steiger	0:7c97dcef700c	710	// here is skipping white space.
steiger	0:7c97dcef700c	711	if ( !p \|\| !*p )
steiger	0:7c97dcef700c	712	{
steiger	0:7c97dcef700c	713	SetError( TIXML_ERROR_DOCUMENT_EMPTY, 0, 0, TIXML_ENCODING_UNKNOWN );
steiger	0:7c97dcef700c	714	return 0;
steiger	0:7c97dcef700c	715	}
steiger	0:7c97dcef700c	716
steiger	0:7c97dcef700c	717	// Note that, for a document, this needs to come
steiger	0:7c97dcef700c	718	// before the while space skip, so that parsing
steiger	0:7c97dcef700c	719	// starts from the pointer we are given.
steiger	0:7c97dcef700c	720	location.Clear();
steiger	0:7c97dcef700c	721	if ( prevData )
steiger	0:7c97dcef700c	722	{
steiger	0:7c97dcef700c	723	location.row = prevData->cursor.row;
steiger	0:7c97dcef700c	724	location.col = prevData->cursor.col;
steiger	0:7c97dcef700c	725	}
steiger	0:7c97dcef700c	726	else
steiger	0:7c97dcef700c	727	{
steiger	0:7c97dcef700c	728	location.row = 0;
steiger	0:7c97dcef700c	729	location.col = 0;
steiger	0:7c97dcef700c	730	}
steiger	0:7c97dcef700c	731	TiXmlParsingData data( p, TabSize(), location.row, location.col );
steiger	0:7c97dcef700c	732	location = data.Cursor();
steiger	0:7c97dcef700c	733
steiger	0:7c97dcef700c	734	if ( encoding == TIXML_ENCODING_UNKNOWN )
steiger	0:7c97dcef700c	735	{
steiger	0:7c97dcef700c	736	// Check for the Microsoft UTF-8 lead bytes.
steiger	0:7c97dcef700c	737	const unsigned char* pU = (const unsigned char*)p;
steiger	0:7c97dcef700c	738	if ( (pU+0) && (pU+0) == TIXML_UTF_LEAD_0
steiger	0:7c97dcef700c	739	&& (pU+1) && (pU+1) == TIXML_UTF_LEAD_1
steiger	0:7c97dcef700c	740	&& (pU+2) && (pU+2) == TIXML_UTF_LEAD_2 )
steiger	0:7c97dcef700c	741	{
steiger	0:7c97dcef700c	742	encoding = TIXML_ENCODING_UTF8;
steiger	0:7c97dcef700c	743	useMicrosoftBOM = true;
steiger	0:7c97dcef700c	744	}
steiger	0:7c97dcef700c	745	}
steiger	0:7c97dcef700c	746
steiger	0:7c97dcef700c	747	p = SkipWhiteSpace( p, encoding );
steiger	0:7c97dcef700c	748	if ( !p )
steiger	0:7c97dcef700c	749	{
steiger	0:7c97dcef700c	750	SetError( TIXML_ERROR_DOCUMENT_EMPTY, 0, 0, TIXML_ENCODING_UNKNOWN );
steiger	0:7c97dcef700c	751	return 0;
steiger	0:7c97dcef700c	752	}
steiger	0:7c97dcef700c	753
steiger	0:7c97dcef700c	754	while ( p && *p )
steiger	0:7c97dcef700c	755	{
steiger	0:7c97dcef700c	756	TiXmlNode* node = Identify( p, encoding );
steiger	0:7c97dcef700c	757	if ( node )
steiger	0:7c97dcef700c	758	{
steiger	0:7c97dcef700c	759	p = node->Parse( p, &data, encoding );
steiger	0:7c97dcef700c	760	LinkEndChild( node );
steiger	0:7c97dcef700c	761	}
steiger	0:7c97dcef700c	762	else
steiger	0:7c97dcef700c	763	{
steiger	0:7c97dcef700c	764	break;
steiger	0:7c97dcef700c	765	}
steiger	0:7c97dcef700c	766
steiger	0:7c97dcef700c	767	// Did we get encoding info?
steiger	0:7c97dcef700c	768	if ( encoding == TIXML_ENCODING_UNKNOWN
steiger	0:7c97dcef700c	769	&& node->ToDeclaration() )
steiger	0:7c97dcef700c	770	{
steiger	0:7c97dcef700c	771	TiXmlDeclaration* dec = node->ToDeclaration();
steiger	0:7c97dcef700c	772	const char* enc = dec->Encoding();
steiger	0:7c97dcef700c	773	assert( enc );
steiger	0:7c97dcef700c	774
steiger	0:7c97dcef700c	775	if ( *enc == 0 )
steiger	0:7c97dcef700c	776	encoding = TIXML_ENCODING_UTF8;
steiger	0:7c97dcef700c	777	else if ( StringEqual( enc, "UTF-8", true, TIXML_ENCODING_UNKNOWN ) )
steiger	0:7c97dcef700c	778	encoding = TIXML_ENCODING_UTF8;
steiger	0:7c97dcef700c	779	else if ( StringEqual( enc, "UTF8", true, TIXML_ENCODING_UNKNOWN ) )
steiger	0:7c97dcef700c	780	encoding = TIXML_ENCODING_UTF8; // incorrect, but be nice
steiger	0:7c97dcef700c	781	else
steiger	0:7c97dcef700c	782	encoding = TIXML_ENCODING_LEGACY;
steiger	0:7c97dcef700c	783	}
steiger	0:7c97dcef700c	784
steiger	0:7c97dcef700c	785	p = SkipWhiteSpace( p, encoding );
steiger	0:7c97dcef700c	786	}
steiger	0:7c97dcef700c	787
steiger	0:7c97dcef700c	788	// Was this empty?
steiger	0:7c97dcef700c	789	if ( !firstChild ) {
steiger	0:7c97dcef700c	790	SetError( TIXML_ERROR_DOCUMENT_EMPTY, 0, 0, encoding );
steiger	0:7c97dcef700c	791	return 0;
steiger	0:7c97dcef700c	792	}
steiger	0:7c97dcef700c	793
steiger	0:7c97dcef700c	794	// All is well.
steiger	0:7c97dcef700c	795	return p;
steiger	0:7c97dcef700c	796	}
steiger	0:7c97dcef700c	797
steiger	0:7c97dcef700c	798	void TiXmlDocument::SetError( int err, const char* pError, TiXmlParsingData* data, TiXmlEncoding encoding )
steiger	0:7c97dcef700c	799	{
steiger	0:7c97dcef700c	800	// The first error in a chain is more accurate - don't set again!
steiger	0:7c97dcef700c	801	if ( error )
steiger	0:7c97dcef700c	802	return;
steiger	0:7c97dcef700c	803
steiger	0:7c97dcef700c	804	assert( err > 0 && err < TIXML_ERROR_STRING_COUNT );
steiger	0:7c97dcef700c	805	error = true;
steiger	0:7c97dcef700c	806	errorId = err;
steiger	0:7c97dcef700c	807	errorDesc = errorString[ errorId ];
steiger	0:7c97dcef700c	808
steiger	0:7c97dcef700c	809	errorLocation.Clear();
steiger	0:7c97dcef700c	810	if ( pError && data )
steiger	0:7c97dcef700c	811	{
steiger	0:7c97dcef700c	812	data->Stamp( pError, encoding );
steiger	0:7c97dcef700c	813	errorLocation = data->Cursor();
steiger	0:7c97dcef700c	814	}
steiger	0:7c97dcef700c	815	}
steiger	0:7c97dcef700c	816
steiger	0:7c97dcef700c	817
steiger	0:7c97dcef700c	818	TiXmlNode* TiXmlNode::Identify( const char* p, TiXmlEncoding encoding )
steiger	0:7c97dcef700c	819	{
steiger	0:7c97dcef700c	820	TiXmlNode* returnNode = 0;
steiger	0:7c97dcef700c	821
steiger	0:7c97dcef700c	822	p = SkipWhiteSpace( p, encoding );
steiger	0:7c97dcef700c	823	if( !p \|\| !p \|\| p != '<' )
steiger	0:7c97dcef700c	824	{
steiger	0:7c97dcef700c	825	return 0;
steiger	0:7c97dcef700c	826	}
steiger	0:7c97dcef700c	827
steiger	0:7c97dcef700c	828	p = SkipWhiteSpace( p, encoding );
steiger	0:7c97dcef700c	829
steiger	0:7c97dcef700c	830	if ( !p \|\| !*p )
steiger	0:7c97dcef700c	831	{
steiger	0:7c97dcef700c	832	return 0;
steiger	0:7c97dcef700c	833	}
steiger	0:7c97dcef700c	834
steiger	0:7c97dcef700c	835	// What is this thing?
steiger	0:7c97dcef700c	836	// - Elements start with a letter or underscore, but xml is reserved.
steiger	0:7c97dcef700c	837	// - Comments: <!--
steiger	0:7c97dcef700c	838	// - Decleration: <?xml
steiger	0:7c97dcef700c	839	// - Everthing else is unknown to tinyxml.
steiger	0:7c97dcef700c	840	//
steiger	0:7c97dcef700c	841
steiger	0:7c97dcef700c	842	const char* xmlHeader = { "<?xml" };
steiger	0:7c97dcef700c	843	const char* commentHeader = { "<!--" };
steiger	0:7c97dcef700c	844	const char* dtdHeader = { "<!" };
steiger	0:7c97dcef700c	845	const char* cdataHeader = { "<![CDATA[" };
steiger	0:7c97dcef700c	846
steiger	0:7c97dcef700c	847	if ( StringEqual( p, xmlHeader, true, encoding ) )
steiger	0:7c97dcef700c	848	{
steiger	0:7c97dcef700c	849	#ifdef DEBUG_PARSER
steiger	0:7c97dcef700c	850	TIXML_LOG( "XML parsing Declaration\n" );
steiger	0:7c97dcef700c	851	#endif
steiger	0:7c97dcef700c	852	returnNode = new TiXmlDeclaration();
steiger	0:7c97dcef700c	853	}
steiger	0:7c97dcef700c	854	else if ( StringEqual( p, commentHeader, false, encoding ) )
steiger	0:7c97dcef700c	855	{
steiger	0:7c97dcef700c	856	#ifdef DEBUG_PARSER
steiger	0:7c97dcef700c	857	TIXML_LOG( "XML parsing Comment\n" );
steiger	0:7c97dcef700c	858	#endif
steiger	0:7c97dcef700c	859	returnNode = new TiXmlComment();
steiger	0:7c97dcef700c	860	}
steiger	0:7c97dcef700c	861	else if ( StringEqual( p, cdataHeader, false, encoding ) )
steiger	0:7c97dcef700c	862	{
steiger	0:7c97dcef700c	863	#ifdef DEBUG_PARSER
steiger	0:7c97dcef700c	864	TIXML_LOG( "XML parsing CDATA\n" );
steiger	0:7c97dcef700c	865	#endif
steiger	0:7c97dcef700c	866	TiXmlText* text = new TiXmlText( "" );
steiger	0:7c97dcef700c	867	text->SetCDATA( true );
steiger	0:7c97dcef700c	868	returnNode = text;
steiger	0:7c97dcef700c	869	}
steiger	0:7c97dcef700c	870	else if ( StringEqual( p, dtdHeader, false, encoding ) )
steiger	0:7c97dcef700c	871	{
steiger	0:7c97dcef700c	872	#ifdef DEBUG_PARSER
steiger	0:7c97dcef700c	873	TIXML_LOG( "XML parsing Unknown(1)\n" );
steiger	0:7c97dcef700c	874	#endif
steiger	0:7c97dcef700c	875	returnNode = new TiXmlUnknown();
steiger	0:7c97dcef700c	876	}
steiger	0:7c97dcef700c	877	else if ( IsAlpha( *(p+1), encoding )
steiger	0:7c97dcef700c	878	\|\| *(p+1) == '_' )
steiger	0:7c97dcef700c	879	{
steiger	0:7c97dcef700c	880	#ifdef DEBUG_PARSER
steiger	0:7c97dcef700c	881	TIXML_LOG( "XML parsing Element\n" );
steiger	0:7c97dcef700c	882	#endif
steiger	0:7c97dcef700c	883	returnNode = new TiXmlElement( "" );
steiger	0:7c97dcef700c	884	}
steiger	0:7c97dcef700c	885	else
steiger	0:7c97dcef700c	886	{
steiger	0:7c97dcef700c	887	#ifdef DEBUG_PARSER
steiger	0:7c97dcef700c	888	TIXML_LOG( "XML parsing Unknown(2)\n" );
steiger	0:7c97dcef700c	889	#endif
steiger	0:7c97dcef700c	890	returnNode = new TiXmlUnknown();
steiger	0:7c97dcef700c	891	}
steiger	0:7c97dcef700c	892
steiger	0:7c97dcef700c	893	if ( returnNode )
steiger	0:7c97dcef700c	894	{
steiger	0:7c97dcef700c	895	// Set the parent, so it can report errors
steiger	0:7c97dcef700c	896	returnNode->parent = this;
steiger	0:7c97dcef700c	897	}
steiger	0:7c97dcef700c	898	return returnNode;
steiger	0:7c97dcef700c	899	}
steiger	0:7c97dcef700c	900
steiger	0:7c97dcef700c	901	#ifdef TIXML_USE_STL
steiger	0:7c97dcef700c	902
steiger	0:7c97dcef700c	903	void TiXmlElement::StreamIn (std::istream * in, TIXML_STRING * tag)
steiger	0:7c97dcef700c	904	{
steiger	0:7c97dcef700c	905	// We're called with some amount of pre-parsing. That is, some of "this"
steiger	0:7c97dcef700c	906	// element is in "tag". Go ahead and stream to the closing ">"
steiger	0:7c97dcef700c	907	while( in->good() )
steiger	0:7c97dcef700c	908	{
steiger	0:7c97dcef700c	909	int c = in->get();
steiger	0:7c97dcef700c	910	if ( c <= 0 )
steiger	0:7c97dcef700c	911	{
steiger	0:7c97dcef700c	912	TiXmlDocument* document = GetDocument();
steiger	0:7c97dcef700c	913	if ( document )
steiger	0:7c97dcef700c	914	document->SetError( TIXML_ERROR_EMBEDDED_NULL, 0, 0, TIXML_ENCODING_UNKNOWN );
steiger	0:7c97dcef700c	915	return;
steiger	0:7c97dcef700c	916	}
steiger	0:7c97dcef700c	917	(*tag) += (char) c ;
steiger	0:7c97dcef700c	918
steiger	0:7c97dcef700c	919	if ( c == '>' )
steiger	0:7c97dcef700c	920	break;
steiger	0:7c97dcef700c	921	}
steiger	0:7c97dcef700c	922
steiger	0:7c97dcef700c	923	if ( tag->length() < 3 ) return;
steiger	0:7c97dcef700c	924
steiger	0:7c97dcef700c	925	// Okay...if we are a "/>" tag, then we're done. We've read a complete tag.
steiger	0:7c97dcef700c	926	// If not, identify and stream.
steiger	0:7c97dcef700c	927
steiger	0:7c97dcef700c	928	if ( tag->at( tag->length() - 1 ) == '>'
steiger	0:7c97dcef700c	929	&& tag->at( tag->length() - 2 ) == '/' )
steiger	0:7c97dcef700c	930	{
steiger	0:7c97dcef700c	931	// All good!
steiger	0:7c97dcef700c	932	return;
steiger	0:7c97dcef700c	933	}
steiger	0:7c97dcef700c	934	else if ( tag->at( tag->length() - 1 ) == '>' )
steiger	0:7c97dcef700c	935	{
steiger	0:7c97dcef700c	936	// There is more. Could be:
steiger	0:7c97dcef700c	937	// text
steiger	0:7c97dcef700c	938	// cdata text (which looks like another node)
steiger	0:7c97dcef700c	939	// closing tag
steiger	0:7c97dcef700c	940	// another node.
steiger	0:7c97dcef700c	941	for ( ;; )
steiger	0:7c97dcef700c	942	{
steiger	0:7c97dcef700c	943	StreamWhiteSpace( in, tag );
steiger	0:7c97dcef700c	944
steiger	0:7c97dcef700c	945	// Do we have text?
steiger	0:7c97dcef700c	946	if ( in->good() && in->peek() != '<' )
steiger	0:7c97dcef700c	947	{
steiger	0:7c97dcef700c	948	// Yep, text.
steiger	0:7c97dcef700c	949	TiXmlText text( "" );
steiger	0:7c97dcef700c	950	text.StreamIn( in, tag );
steiger	0:7c97dcef700c	951
steiger	0:7c97dcef700c	952	// What follows text is a closing tag or another node.
steiger	0:7c97dcef700c	953	// Go around again and figure it out.
steiger	0:7c97dcef700c	954	continue;
steiger	0:7c97dcef700c	955	}
steiger	0:7c97dcef700c	956
steiger	0:7c97dcef700c	957	// We now have either a closing tag...or another node.
steiger	0:7c97dcef700c	958	// We should be at a "<", regardless.
steiger	0:7c97dcef700c	959	if ( !in->good() ) return;
steiger	0:7c97dcef700c	960	assert( in->peek() == '<' );
steiger	0:7c97dcef700c	961	int tagIndex = (int) tag->length();
steiger	0:7c97dcef700c	962
steiger	0:7c97dcef700c	963	bool closingTag = false;
steiger	0:7c97dcef700c	964	bool firstCharFound = false;
steiger	0:7c97dcef700c	965
steiger	0:7c97dcef700c	966	for( ;; )
steiger	0:7c97dcef700c	967	{
steiger	0:7c97dcef700c	968	if ( !in->good() )
steiger	0:7c97dcef700c	969	return;
steiger	0:7c97dcef700c	970
steiger	0:7c97dcef700c	971	int c = in->peek();
steiger	0:7c97dcef700c	972	if ( c <= 0 )
steiger	0:7c97dcef700c	973	{
steiger	0:7c97dcef700c	974	TiXmlDocument* document = GetDocument();
steiger	0:7c97dcef700c	975	if ( document )
steiger	0:7c97dcef700c	976	document->SetError( TIXML_ERROR_EMBEDDED_NULL, 0, 0, TIXML_ENCODING_UNKNOWN );
steiger	0:7c97dcef700c	977	return;
steiger	0:7c97dcef700c	978	}
steiger	0:7c97dcef700c	979
steiger	0:7c97dcef700c	980	if ( c == '>' )
steiger	0:7c97dcef700c	981	break;
steiger	0:7c97dcef700c	982
steiger	0:7c97dcef700c	983	*tag += (char) c;
steiger	0:7c97dcef700c	984	in->get();
steiger	0:7c97dcef700c	985
steiger	0:7c97dcef700c	986	// Early out if we find the CDATA id.
steiger	0:7c97dcef700c	987	if ( c == '[' && tag->size() >= 9 )
steiger	0:7c97dcef700c	988	{
steiger	0:7c97dcef700c	989	size_t len = tag->size();
steiger	0:7c97dcef700c	990	const char* start = tag->c_str() + len - 9;
steiger	0:7c97dcef700c	991	if ( strcmp( start, "<![CDATA[" ) == 0 ) {
steiger	0:7c97dcef700c	992	assert( !closingTag );
steiger	0:7c97dcef700c	993	break;
steiger	0:7c97dcef700c	994	}
steiger	0:7c97dcef700c	995	}
steiger	0:7c97dcef700c	996
steiger	0:7c97dcef700c	997	if ( !firstCharFound && c != '<' && !IsWhiteSpace( c ) )
steiger	0:7c97dcef700c	998	{
steiger	0:7c97dcef700c	999	firstCharFound = true;
steiger	0:7c97dcef700c	1000	if ( c == '/' )
steiger	0:7c97dcef700c	1001	closingTag = true;
steiger	0:7c97dcef700c	1002	}
steiger	0:7c97dcef700c	1003	}
steiger	0:7c97dcef700c	1004	// If it was a closing tag, then read in the closing '>' to clean up the input stream.
steiger	0:7c97dcef700c	1005	// If it was not, the streaming will be done by the tag.
steiger	0:7c97dcef700c	1006	if ( closingTag )
steiger	0:7c97dcef700c	1007	{
steiger	0:7c97dcef700c	1008	if ( !in->good() )
steiger	0:7c97dcef700c	1009	return;
steiger	0:7c97dcef700c	1010
steiger	0:7c97dcef700c	1011	int c = in->get();
steiger	0:7c97dcef700c	1012	if ( c <= 0 )
steiger	0:7c97dcef700c	1013	{
steiger	0:7c97dcef700c	1014	TiXmlDocument* document = GetDocument();
steiger	0:7c97dcef700c	1015	if ( document )
steiger	0:7c97dcef700c	1016	document->SetError( TIXML_ERROR_EMBEDDED_NULL, 0, 0, TIXML_ENCODING_UNKNOWN );
steiger	0:7c97dcef700c	1017	return;
steiger	0:7c97dcef700c	1018	}
steiger	0:7c97dcef700c	1019	assert( c == '>' );
steiger	0:7c97dcef700c	1020	*tag += (char) c;
steiger	0:7c97dcef700c	1021
steiger	0:7c97dcef700c	1022	// We are done, once we've found our closing tag.
steiger	0:7c97dcef700c	1023	return;
steiger	0:7c97dcef700c	1024	}
steiger	0:7c97dcef700c	1025	else
steiger	0:7c97dcef700c	1026	{
steiger	0:7c97dcef700c	1027	// If not a closing tag, id it, and stream.
steiger	0:7c97dcef700c	1028	const char* tagloc = tag->c_str() + tagIndex;
steiger	0:7c97dcef700c	1029	TiXmlNode* node = Identify( tagloc, TIXML_DEFAULT_ENCODING );
steiger	0:7c97dcef700c	1030	if ( !node )
steiger	0:7c97dcef700c	1031	return;
steiger	0:7c97dcef700c	1032	node->StreamIn( in, tag );
steiger	0:7c97dcef700c	1033	delete node;
steiger	0:7c97dcef700c	1034	node = 0;
steiger	0:7c97dcef700c	1035
steiger	0:7c97dcef700c	1036	// No return: go around from the beginning: text, closing tag, or node.
steiger	0:7c97dcef700c	1037	}
steiger	0:7c97dcef700c	1038	}
steiger	0:7c97dcef700c	1039	}
steiger	0:7c97dcef700c	1040	}
steiger	0:7c97dcef700c	1041	#endif
steiger	0:7c97dcef700c	1042
steiger	0:7c97dcef700c	1043	const char* TiXmlElement::Parse( const char* p, TiXmlParsingData* data, TiXmlEncoding encoding )
steiger	0:7c97dcef700c	1044	{
steiger	0:7c97dcef700c	1045	p = SkipWhiteSpace( p, encoding );
steiger	0:7c97dcef700c	1046	TiXmlDocument* document = GetDocument();
steiger	0:7c97dcef700c	1047
steiger	0:7c97dcef700c	1048	if ( !p \|\| !*p )
steiger	0:7c97dcef700c	1049	{
steiger	0:7c97dcef700c	1050	if ( document ) document->SetError( TIXML_ERROR_PARSING_ELEMENT, 0, 0, encoding );
steiger	0:7c97dcef700c	1051	return 0;
steiger	0:7c97dcef700c	1052	}
steiger	0:7c97dcef700c	1053
steiger	0:7c97dcef700c	1054	if ( data )
steiger	0:7c97dcef700c	1055	{
steiger	0:7c97dcef700c	1056	data->Stamp( p, encoding );
steiger	0:7c97dcef700c	1057	location = data->Cursor();
steiger	0:7c97dcef700c	1058	}
steiger	0:7c97dcef700c	1059
steiger	0:7c97dcef700c	1060	if ( *p != '<' )
steiger	0:7c97dcef700c	1061	{
steiger	0:7c97dcef700c	1062	if ( document ) document->SetError( TIXML_ERROR_PARSING_ELEMENT, p, data, encoding );
steiger	0:7c97dcef700c	1063	return 0;
steiger	0:7c97dcef700c	1064	}
steiger	0:7c97dcef700c	1065
steiger	0:7c97dcef700c	1066	p = SkipWhiteSpace( p+1, encoding );
steiger	0:7c97dcef700c	1067
steiger	0:7c97dcef700c	1068	// Read the name.
steiger	0:7c97dcef700c	1069	const char* pErr = p;
steiger	0:7c97dcef700c	1070
steiger	0:7c97dcef700c	1071	p = ReadName( p, &value, encoding );
steiger	0:7c97dcef700c	1072	if ( !p \|\| !*p )
steiger	0:7c97dcef700c	1073	{
steiger	0:7c97dcef700c	1074	if ( document ) document->SetError( TIXML_ERROR_FAILED_TO_READ_ELEMENT_NAME, pErr, data, encoding );
steiger	0:7c97dcef700c	1075	return 0;
steiger	0:7c97dcef700c	1076	}
steiger	0:7c97dcef700c	1077
steiger	0:7c97dcef700c	1078	TIXML_STRING endTag ("</");
steiger	0:7c97dcef700c	1079	endTag += value;
steiger	0:7c97dcef700c	1080
steiger	0:7c97dcef700c	1081	// Check for and read attributes. Also look for an empty
steiger	0:7c97dcef700c	1082	// tag or an end tag.
steiger	0:7c97dcef700c	1083	while ( p && *p )
steiger	0:7c97dcef700c	1084	{
steiger	0:7c97dcef700c	1085	pErr = p;
steiger	0:7c97dcef700c	1086	p = SkipWhiteSpace( p, encoding );
steiger	0:7c97dcef700c	1087	if ( !p \|\| !*p )
steiger	0:7c97dcef700c	1088	{
steiger	0:7c97dcef700c	1089	if ( document ) document->SetError( TIXML_ERROR_READING_ATTRIBUTES, pErr, data, encoding );
steiger	0:7c97dcef700c	1090	return 0;
steiger	0:7c97dcef700c	1091	}
steiger	0:7c97dcef700c	1092	if ( *p == '/' )
steiger	0:7c97dcef700c	1093	{
steiger	0:7c97dcef700c	1094	++p;
steiger	0:7c97dcef700c	1095	// Empty tag.
steiger	0:7c97dcef700c	1096	if ( *p != '>' )
steiger	0:7c97dcef700c	1097	{
steiger	0:7c97dcef700c	1098	if ( document ) document->SetError( TIXML_ERROR_PARSING_EMPTY, p, data, encoding );
steiger	0:7c97dcef700c	1099	return 0;
steiger	0:7c97dcef700c	1100	}
steiger	0:7c97dcef700c	1101	return (p+1);
steiger	0:7c97dcef700c	1102	}
steiger	0:7c97dcef700c	1103	else if ( *p == '>' )
steiger	0:7c97dcef700c	1104	{
steiger	0:7c97dcef700c	1105	// Done with attributes (if there were any.)
steiger	0:7c97dcef700c	1106	// Read the value -- which can include other
steiger	0:7c97dcef700c	1107	// elements -- read the end tag, and return.
steiger	0:7c97dcef700c	1108	++p;
steiger	0:7c97dcef700c	1109	p = ReadValue( p, data, encoding ); // Note this is an Element method, and will set the error if one happens.
steiger	0:7c97dcef700c	1110	if ( !p \|\| !*p ) {
steiger	0:7c97dcef700c	1111	// We were looking for the end tag, but found nothing.
steiger	0:7c97dcef700c	1112	// Fix for [ 1663758 ] Failure to report error on bad XML
steiger	0:7c97dcef700c	1113	if ( document ) document->SetError( TIXML_ERROR_READING_END_TAG, p, data, encoding );
steiger	0:7c97dcef700c	1114	return 0;
steiger	0:7c97dcef700c	1115	}
steiger	0:7c97dcef700c	1116
steiger	0:7c97dcef700c	1117	// We should find the end tag now
steiger	0:7c97dcef700c	1118	// note that:
steiger	0:7c97dcef700c	1119	// </foo > and
steiger	0:7c97dcef700c	1120	// </foo>
steiger	0:7c97dcef700c	1121	// are both valid end tags.
steiger	0:7c97dcef700c	1122	if ( StringEqual( p, endTag.c_str(), false, encoding ) )
steiger	0:7c97dcef700c	1123	{
steiger	0:7c97dcef700c	1124	p += endTag.length();
steiger	0:7c97dcef700c	1125	p = SkipWhiteSpace( p, encoding );
steiger	0:7c97dcef700c	1126	if ( p && p && p == '>' ) {
steiger	0:7c97dcef700c	1127	++p;
steiger	0:7c97dcef700c	1128	return p;
steiger	0:7c97dcef700c	1129	}
steiger	0:7c97dcef700c	1130	if ( document ) document->SetError( TIXML_ERROR_READING_END_TAG, p, data, encoding );
steiger	0:7c97dcef700c	1131	return 0;
steiger	0:7c97dcef700c	1132	}
steiger	0:7c97dcef700c	1133	else
steiger	0:7c97dcef700c	1134	{
steiger	0:7c97dcef700c	1135	if ( document ) document->SetError( TIXML_ERROR_READING_END_TAG, p, data, encoding );
steiger	0:7c97dcef700c	1136	return 0;
steiger	0:7c97dcef700c	1137	}
steiger	0:7c97dcef700c	1138	}
steiger	0:7c97dcef700c	1139	else
steiger	0:7c97dcef700c	1140	{
steiger	0:7c97dcef700c	1141	// Try to read an attribute:
steiger	0:7c97dcef700c	1142	TiXmlAttribute* attrib = new TiXmlAttribute();
steiger	0:7c97dcef700c	1143	if ( !attrib )
steiger	0:7c97dcef700c	1144	{
steiger	0:7c97dcef700c	1145	return 0;
steiger	0:7c97dcef700c	1146	}
steiger	0:7c97dcef700c	1147
steiger	0:7c97dcef700c	1148	attrib->SetDocument( document );
steiger	0:7c97dcef700c	1149	pErr = p;
steiger	0:7c97dcef700c	1150	p = attrib->Parse( p, data, encoding );
steiger	0:7c97dcef700c	1151
steiger	0:7c97dcef700c	1152	if ( !p \|\| !*p )
steiger	0:7c97dcef700c	1153	{
steiger	0:7c97dcef700c	1154	if ( document ) document->SetError( TIXML_ERROR_PARSING_ELEMENT, pErr, data, encoding );
steiger	0:7c97dcef700c	1155	delete attrib;
steiger	0:7c97dcef700c	1156	return 0;
steiger	0:7c97dcef700c	1157	}
steiger	0:7c97dcef700c	1158
steiger	0:7c97dcef700c	1159	// Handle the strange case of double attributes:
steiger	0:7c97dcef700c	1160	#ifdef TIXML_USE_STL
steiger	0:7c97dcef700c	1161	TiXmlAttribute* node = attributeSet.Find( attrib->NameTStr() );
steiger	0:7c97dcef700c	1162	#else
steiger	0:7c97dcef700c	1163	TiXmlAttribute* node = attributeSet.Find( attrib->Name() );
steiger	0:7c97dcef700c	1164	#endif
steiger	0:7c97dcef700c	1165	if ( node )
steiger	0:7c97dcef700c	1166	{
steiger	0:7c97dcef700c	1167	if ( document ) document->SetError( TIXML_ERROR_PARSING_ELEMENT, pErr, data, encoding );
steiger	0:7c97dcef700c	1168	delete attrib;
steiger	0:7c97dcef700c	1169	return 0;
steiger	0:7c97dcef700c	1170	}
steiger	0:7c97dcef700c	1171
steiger	0:7c97dcef700c	1172	attributeSet.Add( attrib );
steiger	0:7c97dcef700c	1173	}
steiger	0:7c97dcef700c	1174	}
steiger	0:7c97dcef700c	1175	return p;
steiger	0:7c97dcef700c	1176	}
steiger	0:7c97dcef700c	1177
steiger	0:7c97dcef700c	1178
steiger	0:7c97dcef700c	1179	const char* TiXmlElement::ReadValue( const char* p, TiXmlParsingData* data, TiXmlEncoding encoding )
steiger	0:7c97dcef700c	1180	{
steiger	0:7c97dcef700c	1181	TiXmlDocument* document = GetDocument();
steiger	0:7c97dcef700c	1182
steiger	0:7c97dcef700c	1183	// Read in text and elements in any order.
steiger	0:7c97dcef700c	1184	const char* pWithWhiteSpace = p;
steiger	0:7c97dcef700c	1185	p = SkipWhiteSpace( p, encoding );
steiger	0:7c97dcef700c	1186
steiger	0:7c97dcef700c	1187	while ( p && *p )
steiger	0:7c97dcef700c	1188	{
steiger	0:7c97dcef700c	1189	if ( *p != '<' )
steiger	0:7c97dcef700c	1190	{
steiger	0:7c97dcef700c	1191	// Take what we have, make a text element.
steiger	0:7c97dcef700c	1192	TiXmlText* textNode = new TiXmlText( "" );
steiger	0:7c97dcef700c	1193
steiger	0:7c97dcef700c	1194	if ( !textNode )
steiger	0:7c97dcef700c	1195	{
steiger	0:7c97dcef700c	1196	return 0;
steiger	0:7c97dcef700c	1197	}
steiger	0:7c97dcef700c	1198
steiger	0:7c97dcef700c	1199	if ( TiXmlBase::IsWhiteSpaceCondensed() )
steiger	0:7c97dcef700c	1200	{
steiger	0:7c97dcef700c	1201	p = textNode->Parse( p, data, encoding );
steiger	0:7c97dcef700c	1202	}
steiger	0:7c97dcef700c	1203	else
steiger	0:7c97dcef700c	1204	{
steiger	0:7c97dcef700c	1205	// Special case: we want to keep the white space
steiger	0:7c97dcef700c	1206	// so that leading spaces aren't removed.
steiger	0:7c97dcef700c	1207	p = textNode->Parse( pWithWhiteSpace, data, encoding );
steiger	0:7c97dcef700c	1208	}
steiger	0:7c97dcef700c	1209
steiger	0:7c97dcef700c	1210	if ( !textNode->Blank() )
steiger	0:7c97dcef700c	1211	LinkEndChild( textNode );
steiger	0:7c97dcef700c	1212	else
steiger	0:7c97dcef700c	1213	delete textNode;
steiger	0:7c97dcef700c	1214	}
steiger	0:7c97dcef700c	1215	else
steiger	0:7c97dcef700c	1216	{
steiger	0:7c97dcef700c	1217	// We hit a '<'
steiger	0:7c97dcef700c	1218	// Have we hit a new element or an end tag? This could also be
steiger	0:7c97dcef700c	1219	// a TiXmlText in the "CDATA" style.
steiger	0:7c97dcef700c	1220	if ( StringEqual( p, "</", false, encoding ) )
steiger	0:7c97dcef700c	1221	{
steiger	0:7c97dcef700c	1222	return p;
steiger	0:7c97dcef700c	1223	}
steiger	0:7c97dcef700c	1224	else
steiger	0:7c97dcef700c	1225	{
steiger	0:7c97dcef700c	1226	TiXmlNode* node = Identify( p, encoding );
steiger	0:7c97dcef700c	1227	if ( node )
steiger	0:7c97dcef700c	1228	{
steiger	0:7c97dcef700c	1229	p = node->Parse( p, data, encoding );
steiger	0:7c97dcef700c	1230	LinkEndChild( node );
steiger	0:7c97dcef700c	1231	}
steiger	0:7c97dcef700c	1232	else
steiger	0:7c97dcef700c	1233	{
steiger	0:7c97dcef700c	1234	return 0;
steiger	0:7c97dcef700c	1235	}
steiger	0:7c97dcef700c	1236	}
steiger	0:7c97dcef700c	1237	}
steiger	0:7c97dcef700c	1238	pWithWhiteSpace = p;
steiger	0:7c97dcef700c	1239	p = SkipWhiteSpace( p, encoding );
steiger	0:7c97dcef700c	1240	}
steiger	0:7c97dcef700c	1241
steiger	0:7c97dcef700c	1242	if ( !p )
steiger	0:7c97dcef700c	1243	{
steiger	0:7c97dcef700c	1244	if ( document ) document->SetError( TIXML_ERROR_READING_ELEMENT_VALUE, 0, 0, encoding );
steiger	0:7c97dcef700c	1245	}
steiger	0:7c97dcef700c	1246	return p;
steiger	0:7c97dcef700c	1247	}
steiger	0:7c97dcef700c	1248
steiger	0:7c97dcef700c	1249
steiger	0:7c97dcef700c	1250	#ifdef TIXML_USE_STL
steiger	0:7c97dcef700c	1251	void TiXmlUnknown::StreamIn( std::istream * in, TIXML_STRING * tag )
steiger	0:7c97dcef700c	1252	{
steiger	0:7c97dcef700c	1253	while ( in->good() )
steiger	0:7c97dcef700c	1254	{
steiger	0:7c97dcef700c	1255	int c = in->get();
steiger	0:7c97dcef700c	1256	if ( c <= 0 )
steiger	0:7c97dcef700c	1257	{
steiger	0:7c97dcef700c	1258	TiXmlDocument* document = GetDocument();
steiger	0:7c97dcef700c	1259	if ( document )
steiger	0:7c97dcef700c	1260	document->SetError( TIXML_ERROR_EMBEDDED_NULL, 0, 0, TIXML_ENCODING_UNKNOWN );
steiger	0:7c97dcef700c	1261	return;
steiger	0:7c97dcef700c	1262	}
steiger	0:7c97dcef700c	1263	(*tag) += (char) c;
steiger	0:7c97dcef700c	1264
steiger	0:7c97dcef700c	1265	if ( c == '>' )
steiger	0:7c97dcef700c	1266	{
steiger	0:7c97dcef700c	1267	// All is well.
steiger	0:7c97dcef700c	1268	return;
steiger	0:7c97dcef700c	1269	}
steiger	0:7c97dcef700c	1270	}
steiger	0:7c97dcef700c	1271	}
steiger	0:7c97dcef700c	1272	#endif
steiger	0:7c97dcef700c	1273
steiger	0:7c97dcef700c	1274
steiger	0:7c97dcef700c	1275	const char* TiXmlUnknown::Parse( const char* p, TiXmlParsingData* data, TiXmlEncoding encoding )
steiger	0:7c97dcef700c	1276	{
steiger	0:7c97dcef700c	1277	TiXmlDocument* document = GetDocument();
steiger	0:7c97dcef700c	1278	p = SkipWhiteSpace( p, encoding );
steiger	0:7c97dcef700c	1279
steiger	0:7c97dcef700c	1280	if ( data )
steiger	0:7c97dcef700c	1281	{
steiger	0:7c97dcef700c	1282	data->Stamp( p, encoding );
steiger	0:7c97dcef700c	1283	location = data->Cursor();
steiger	0:7c97dcef700c	1284	}
steiger	0:7c97dcef700c	1285	if ( !p \|\| !p \|\| p != '<' )
steiger	0:7c97dcef700c	1286	{
steiger	0:7c97dcef700c	1287	if ( document ) document->SetError( TIXML_ERROR_PARSING_UNKNOWN, p, data, encoding );
steiger	0:7c97dcef700c	1288	return 0;
steiger	0:7c97dcef700c	1289	}
steiger	0:7c97dcef700c	1290	++p;
steiger	0:7c97dcef700c	1291	value = "";
steiger	0:7c97dcef700c	1292
steiger	0:7c97dcef700c	1293	while ( p && p && p != '>' )
steiger	0:7c97dcef700c	1294	{
steiger	0:7c97dcef700c	1295	value += *p;
steiger	0:7c97dcef700c	1296	++p;
steiger	0:7c97dcef700c	1297	}
steiger	0:7c97dcef700c	1298
steiger	0:7c97dcef700c	1299	if ( !p )
steiger	0:7c97dcef700c	1300	{
steiger	0:7c97dcef700c	1301	if ( document ) document->SetError( TIXML_ERROR_PARSING_UNKNOWN, 0, 0, encoding );
steiger	0:7c97dcef700c	1302	}
steiger	0:7c97dcef700c	1303	if ( *p == '>' )
steiger	0:7c97dcef700c	1304	return p+1;
steiger	0:7c97dcef700c	1305	return p;
steiger	0:7c97dcef700c	1306	}
steiger	0:7c97dcef700c	1307
steiger	0:7c97dcef700c	1308	#ifdef TIXML_USE_STL
steiger	0:7c97dcef700c	1309	void TiXmlComment::StreamIn( std::istream * in, TIXML_STRING * tag )
steiger	0:7c97dcef700c	1310	{
steiger	0:7c97dcef700c	1311	while ( in->good() )
steiger	0:7c97dcef700c	1312	{
steiger	0:7c97dcef700c	1313	int c = in->get();
steiger	0:7c97dcef700c	1314	if ( c <= 0 )
steiger	0:7c97dcef700c	1315	{
steiger	0:7c97dcef700c	1316	TiXmlDocument* document = GetDocument();
steiger	0:7c97dcef700c	1317	if ( document )
steiger	0:7c97dcef700c	1318	document->SetError( TIXML_ERROR_EMBEDDED_NULL, 0, 0, TIXML_ENCODING_UNKNOWN );
steiger	0:7c97dcef700c	1319	return;
steiger	0:7c97dcef700c	1320	}
steiger	0:7c97dcef700c	1321
steiger	0:7c97dcef700c	1322	(*tag) += (char) c;
steiger	0:7c97dcef700c	1323
steiger	0:7c97dcef700c	1324	if ( c == '>'
steiger	0:7c97dcef700c	1325	&& tag->at( tag->length() - 2 ) == '-'
steiger	0:7c97dcef700c	1326	&& tag->at( tag->length() - 3 ) == '-' )
steiger	0:7c97dcef700c	1327	{
steiger	0:7c97dcef700c	1328	// All is well.
steiger	0:7c97dcef700c	1329	return;
steiger	0:7c97dcef700c	1330	}
steiger	0:7c97dcef700c	1331	}
steiger	0:7c97dcef700c	1332	}
steiger	0:7c97dcef700c	1333	#endif
steiger	0:7c97dcef700c	1334
steiger	0:7c97dcef700c	1335
steiger	0:7c97dcef700c	1336	const char* TiXmlComment::Parse( const char* p, TiXmlParsingData* data, TiXmlEncoding encoding )
steiger	0:7c97dcef700c	1337	{
steiger	0:7c97dcef700c	1338	TiXmlDocument* document = GetDocument();
steiger	0:7c97dcef700c	1339	value = "";
steiger	0:7c97dcef700c	1340
steiger	0:7c97dcef700c	1341	p = SkipWhiteSpace( p, encoding );
steiger	0:7c97dcef700c	1342
steiger	0:7c97dcef700c	1343	if ( data )
steiger	0:7c97dcef700c	1344	{
steiger	0:7c97dcef700c	1345	data->Stamp( p, encoding );
steiger	0:7c97dcef700c	1346	location = data->Cursor();
steiger	0:7c97dcef700c	1347	}
steiger	0:7c97dcef700c	1348	const char* startTag = "<!--";
steiger	0:7c97dcef700c	1349	const char* endTag = "-->";
steiger	0:7c97dcef700c	1350
steiger	0:7c97dcef700c	1351	if ( !StringEqual( p, startTag, false, encoding ) )
steiger	0:7c97dcef700c	1352	{
steiger	0:7c97dcef700c	1353	document->SetError( TIXML_ERROR_PARSING_COMMENT, p, data, encoding );
steiger	0:7c97dcef700c	1354	return 0;
steiger	0:7c97dcef700c	1355	}
steiger	0:7c97dcef700c	1356	p += strlen( startTag );
steiger	0:7c97dcef700c	1357
steiger	0:7c97dcef700c	1358	// [ 1475201 ] TinyXML parses entities in comments
steiger	0:7c97dcef700c	1359	// Oops - ReadText doesn't work, because we don't want to parse the entities.
steiger	0:7c97dcef700c	1360	// p = ReadText( p, &value, false, endTag, false, encoding );
steiger	0:7c97dcef700c	1361	//
steiger	0:7c97dcef700c	1362	// from the XML spec:
steiger	0:7c97dcef700c	1363	/*
steiger	0:7c97dcef700c	1364	[Definition: Comments may appear anywhere in a document outside other markup; in addition,
steiger	0:7c97dcef700c	1365	they may appear within the document type declaration at places allowed by the grammar.
steiger	0:7c97dcef700c	1366	They are not part of the document's character data; an XML processor MAY, but need not,
steiger	0:7c97dcef700c	1367	make it possible for an application to retrieve the text of comments. For compatibility,
steiger	0:7c97dcef700c	1368	the string "--" (double-hyphen) MUST NOT occur within comments.] Parameter entity
steiger	0:7c97dcef700c	1369	references MUST NOT be recognized within comments.
steiger	0:7c97dcef700c	1370
steiger	0:7c97dcef700c	1371	An example of a comment:
steiger	0:7c97dcef700c	1372
steiger	0:7c97dcef700c	1373	<!-- declarations for <head> & <body> -->
steiger	0:7c97dcef700c	1374	*/
steiger	0:7c97dcef700c	1375
steiger	0:7c97dcef700c	1376	value = "";
steiger	0:7c97dcef700c	1377	// Keep all the white space.
steiger	0:7c97dcef700c	1378	while ( p && *p && !StringEqual( p, endTag, false, encoding ) )
steiger	0:7c97dcef700c	1379	{
steiger	0:7c97dcef700c	1380	value.append( p, 1 );
steiger	0:7c97dcef700c	1381	++p;
steiger	0:7c97dcef700c	1382	}
steiger	0:7c97dcef700c	1383	if ( p && *p )
steiger	0:7c97dcef700c	1384	p += strlen( endTag );
steiger	0:7c97dcef700c	1385
steiger	0:7c97dcef700c	1386	return p;
steiger	0:7c97dcef700c	1387	}
steiger	0:7c97dcef700c	1388
steiger	0:7c97dcef700c	1389
steiger	0:7c97dcef700c	1390	const char* TiXmlAttribute::Parse( const char* p, TiXmlParsingData* data, TiXmlEncoding encoding )
steiger	0:7c97dcef700c	1391	{
steiger	0:7c97dcef700c	1392	p = SkipWhiteSpace( p, encoding );
steiger	0:7c97dcef700c	1393	if ( !p \|\| !*p ) return 0;
steiger	0:7c97dcef700c	1394
steiger	0:7c97dcef700c	1395	if ( data )
steiger	0:7c97dcef700c	1396	{
steiger	0:7c97dcef700c	1397	data->Stamp( p, encoding );
steiger	0:7c97dcef700c	1398	location = data->Cursor();
steiger	0:7c97dcef700c	1399	}
steiger	0:7c97dcef700c	1400	// Read the name, the '=' and the value.
steiger	0:7c97dcef700c	1401	const char* pErr = p;
steiger	0:7c97dcef700c	1402	p = ReadName( p, &name, encoding );
steiger	0:7c97dcef700c	1403	if ( !p \|\| !*p )
steiger	0:7c97dcef700c	1404	{
steiger	0:7c97dcef700c	1405	if ( document ) document->SetError( TIXML_ERROR_READING_ATTRIBUTES, pErr, data, encoding );
steiger	0:7c97dcef700c	1406	return 0;
steiger	0:7c97dcef700c	1407	}
steiger	0:7c97dcef700c	1408	p = SkipWhiteSpace( p, encoding );
steiger	0:7c97dcef700c	1409	if ( !p \|\| !p \|\| p != '=' )
steiger	0:7c97dcef700c	1410	{
steiger	0:7c97dcef700c	1411	if ( document ) document->SetError( TIXML_ERROR_READING_ATTRIBUTES, p, data, encoding );
steiger	0:7c97dcef700c	1412	return 0;
steiger	0:7c97dcef700c	1413	}
steiger	0:7c97dcef700c	1414
steiger	0:7c97dcef700c	1415	++p; // skip '='
steiger	0:7c97dcef700c	1416	p = SkipWhiteSpace( p, encoding );
steiger	0:7c97dcef700c	1417	if ( !p \|\| !*p )
steiger	0:7c97dcef700c	1418	{
steiger	0:7c97dcef700c	1419	if ( document ) document->SetError( TIXML_ERROR_READING_ATTRIBUTES, p, data, encoding );
steiger	0:7c97dcef700c	1420	return 0;
steiger	0:7c97dcef700c	1421	}
steiger	0:7c97dcef700c	1422
steiger	0:7c97dcef700c	1423	const char* end;
steiger	0:7c97dcef700c	1424	const char SINGLE_QUOTE = '\'';
steiger	0:7c97dcef700c	1425	const char DOUBLE_QUOTE = '\"';
steiger	0:7c97dcef700c	1426
steiger	0:7c97dcef700c	1427	if ( *p == SINGLE_QUOTE )
steiger	0:7c97dcef700c	1428	{
steiger	0:7c97dcef700c	1429	++p;
steiger	0:7c97dcef700c	1430	end = "\'"; // single quote in string
steiger	0:7c97dcef700c	1431	p = ReadText( p, &value, false, end, false, encoding );
steiger	0:7c97dcef700c	1432	}
steiger	0:7c97dcef700c	1433	else if ( *p == DOUBLE_QUOTE )
steiger	0:7c97dcef700c	1434	{
steiger	0:7c97dcef700c	1435	++p;
steiger	0:7c97dcef700c	1436	end = "\""; // double quote in string
steiger	0:7c97dcef700c	1437	p = ReadText( p, &value, false, end, false, encoding );
steiger	0:7c97dcef700c	1438	}
steiger	0:7c97dcef700c	1439	else
steiger	0:7c97dcef700c	1440	{
steiger	0:7c97dcef700c	1441	// All attribute values should be in single or double quotes.
steiger	0:7c97dcef700c	1442	// But this is such a common error that the parser will try
steiger	0:7c97dcef700c	1443	// its best, even without them.
steiger	0:7c97dcef700c	1444	value = "";
steiger	0:7c97dcef700c	1445	while ( p && *p // existence
steiger	0:7c97dcef700c	1446	&& !IsWhiteSpace( *p ) // whitespace
steiger	0:7c97dcef700c	1447	&& p != '/' && p != '>' ) // tag end
steiger	0:7c97dcef700c	1448	{
steiger	0:7c97dcef700c	1449	if ( p == SINGLE_QUOTE \|\| p == DOUBLE_QUOTE ) {
steiger	0:7c97dcef700c	1450	// [ 1451649 ] Attribute values with trailing quotes not handled correctly
steiger	0:7c97dcef700c	1451	// We did not have an opening quote but seem to have a
steiger	0:7c97dcef700c	1452	// closing one. Give up and throw an error.
steiger	0:7c97dcef700c	1453	if ( document ) document->SetError( TIXML_ERROR_READING_ATTRIBUTES, p, data, encoding );
steiger	0:7c97dcef700c	1454	return 0;
steiger	0:7c97dcef700c	1455	}
steiger	0:7c97dcef700c	1456	value += *p;
steiger	0:7c97dcef700c	1457	++p;
steiger	0:7c97dcef700c	1458	}
steiger	0:7c97dcef700c	1459	}
steiger	0:7c97dcef700c	1460	return p;
steiger	0:7c97dcef700c	1461	}
steiger	0:7c97dcef700c	1462
steiger	0:7c97dcef700c	1463	#ifdef TIXML_USE_STL
steiger	0:7c97dcef700c	1464	void TiXmlText::StreamIn( std::istream * in, TIXML_STRING * tag )
steiger	0:7c97dcef700c	1465	{
steiger	0:7c97dcef700c	1466	while ( in->good() )
steiger	0:7c97dcef700c	1467	{
steiger	0:7c97dcef700c	1468	int c = in->peek();
steiger	0:7c97dcef700c	1469	if ( !cdata && (c == '<' ) )
steiger	0:7c97dcef700c	1470	{
steiger	0:7c97dcef700c	1471	return;
steiger	0:7c97dcef700c	1472	}
steiger	0:7c97dcef700c	1473	if ( c <= 0 )
steiger	0:7c97dcef700c	1474	{
steiger	0:7c97dcef700c	1475	TiXmlDocument* document = GetDocument();
steiger	0:7c97dcef700c	1476	if ( document )
steiger	0:7c97dcef700c	1477	document->SetError( TIXML_ERROR_EMBEDDED_NULL, 0, 0, TIXML_ENCODING_UNKNOWN );
steiger	0:7c97dcef700c	1478	return;
steiger	0:7c97dcef700c	1479	}
steiger	0:7c97dcef700c	1480
steiger	0:7c97dcef700c	1481	(*tag) += (char) c;
steiger	0:7c97dcef700c	1482	in->get(); // "commits" the peek made above
steiger	0:7c97dcef700c	1483
steiger	0:7c97dcef700c	1484	if ( cdata && c == '>' && tag->size() >= 3 ) {
steiger	0:7c97dcef700c	1485	size_t len = tag->size();
steiger	0:7c97dcef700c	1486	if ( (tag)[len-2] == ']' && (tag)[len-3] == ']' ) {
steiger	0:7c97dcef700c	1487	// terminator of cdata.
steiger	0:7c97dcef700c	1488	return;
steiger	0:7c97dcef700c	1489	}
steiger	0:7c97dcef700c	1490	}
steiger	0:7c97dcef700c	1491	}
steiger	0:7c97dcef700c	1492	}
steiger	0:7c97dcef700c	1493	#endif
steiger	0:7c97dcef700c	1494
steiger	0:7c97dcef700c	1495	const char* TiXmlText::Parse( const char* p, TiXmlParsingData* data, TiXmlEncoding encoding )
steiger	0:7c97dcef700c	1496	{
steiger	0:7c97dcef700c	1497	value = "";
steiger	0:7c97dcef700c	1498	TiXmlDocument* document = GetDocument();
steiger	0:7c97dcef700c	1499
steiger	0:7c97dcef700c	1500	if ( data )
steiger	0:7c97dcef700c	1501	{
steiger	0:7c97dcef700c	1502	data->Stamp( p, encoding );
steiger	0:7c97dcef700c	1503	location = data->Cursor();
steiger	0:7c97dcef700c	1504	}
steiger	0:7c97dcef700c	1505
steiger	0:7c97dcef700c	1506	const char* const startTag = "<![CDATA[";
steiger	0:7c97dcef700c	1507	const char* const endTag = "]]>";
steiger	0:7c97dcef700c	1508
steiger	0:7c97dcef700c	1509	if ( cdata \|\| StringEqual( p, startTag, false, encoding ) )
steiger	0:7c97dcef700c	1510	{
steiger	0:7c97dcef700c	1511	cdata = true;
steiger	0:7c97dcef700c	1512
steiger	0:7c97dcef700c	1513	if ( !StringEqual( p, startTag, false, encoding ) )
steiger	0:7c97dcef700c	1514	{
steiger	0:7c97dcef700c	1515	document->SetError( TIXML_ERROR_PARSING_CDATA, p, data, encoding );
steiger	0:7c97dcef700c	1516	return 0;
steiger	0:7c97dcef700c	1517	}
steiger	0:7c97dcef700c	1518	p += strlen( startTag );
steiger	0:7c97dcef700c	1519
steiger	0:7c97dcef700c	1520	// Keep all the white space, ignore the encoding, etc.
steiger	0:7c97dcef700c	1521	while ( p && *p
steiger	0:7c97dcef700c	1522	&& !StringEqual( p, endTag, false, encoding )
steiger	0:7c97dcef700c	1523	)
steiger	0:7c97dcef700c	1524	{
steiger	0:7c97dcef700c	1525	value += *p;
steiger	0:7c97dcef700c	1526	++p;
steiger	0:7c97dcef700c	1527	}
steiger	0:7c97dcef700c	1528
steiger	0:7c97dcef700c	1529	TIXML_STRING dummy;
steiger	0:7c97dcef700c	1530	p = ReadText( p, &dummy, false, endTag, false, encoding );
steiger	0:7c97dcef700c	1531	return p;
steiger	0:7c97dcef700c	1532	}
steiger	0:7c97dcef700c	1533	else
steiger	0:7c97dcef700c	1534	{
steiger	0:7c97dcef700c	1535	bool ignoreWhite = true;
steiger	0:7c97dcef700c	1536
steiger	0:7c97dcef700c	1537	const char* end = "<";
steiger	0:7c97dcef700c	1538	p = ReadText( p, &value, ignoreWhite, end, false, encoding );
steiger	0:7c97dcef700c	1539	if ( p )
steiger	0:7c97dcef700c	1540	return p-1; // don't truncate the '<'
steiger	0:7c97dcef700c	1541	return 0;
steiger	0:7c97dcef700c	1542	}
steiger	0:7c97dcef700c	1543	}
steiger	0:7c97dcef700c	1544
steiger	0:7c97dcef700c	1545	#ifdef TIXML_USE_STL
steiger	0:7c97dcef700c	1546	void TiXmlDeclaration::StreamIn( std::istream * in, TIXML_STRING * tag )
steiger	0:7c97dcef700c	1547	{
steiger	0:7c97dcef700c	1548	while ( in->good() )
steiger	0:7c97dcef700c	1549	{
steiger	0:7c97dcef700c	1550	int c = in->get();
steiger	0:7c97dcef700c	1551	if ( c <= 0 )
steiger	0:7c97dcef700c	1552	{
steiger	0:7c97dcef700c	1553	TiXmlDocument* document = GetDocument();
steiger	0:7c97dcef700c	1554	if ( document )
steiger	0:7c97dcef700c	1555	document->SetError( TIXML_ERROR_EMBEDDED_NULL, 0, 0, TIXML_ENCODING_UNKNOWN );
steiger	0:7c97dcef700c	1556	return;
steiger	0:7c97dcef700c	1557	}
steiger	0:7c97dcef700c	1558	(*tag) += (char) c;
steiger	0:7c97dcef700c	1559
steiger	0:7c97dcef700c	1560	if ( c == '>' )
steiger	0:7c97dcef700c	1561	{
steiger	0:7c97dcef700c	1562	// All is well.
steiger	0:7c97dcef700c	1563	return;
steiger	0:7c97dcef700c	1564	}
steiger	0:7c97dcef700c	1565	}
steiger	0:7c97dcef700c	1566	}
steiger	0:7c97dcef700c	1567	#endif
steiger	0:7c97dcef700c	1568
steiger	0:7c97dcef700c	1569	const char* TiXmlDeclaration::Parse( const char* p, TiXmlParsingData* data, TiXmlEncoding _encoding )
steiger	0:7c97dcef700c	1570	{
steiger	0:7c97dcef700c	1571	p = SkipWhiteSpace( p, _encoding );
steiger	0:7c97dcef700c	1572	// Find the beginning, find the end, and look for
steiger	0:7c97dcef700c	1573	// the stuff in-between.
steiger	0:7c97dcef700c	1574	TiXmlDocument* document = GetDocument();
steiger	0:7c97dcef700c	1575	if ( !p \|\| !*p \|\| !StringEqual( p, "<?xml", true, _encoding ) )
steiger	0:7c97dcef700c	1576	{
steiger	0:7c97dcef700c	1577	if ( document ) document->SetError( TIXML_ERROR_PARSING_DECLARATION, 0, 0, _encoding );
steiger	0:7c97dcef700c	1578	return 0;
steiger	0:7c97dcef700c	1579	}
steiger	0:7c97dcef700c	1580	if ( data )
steiger	0:7c97dcef700c	1581	{
steiger	0:7c97dcef700c	1582	data->Stamp( p, _encoding );
steiger	0:7c97dcef700c	1583	location = data->Cursor();
steiger	0:7c97dcef700c	1584	}
steiger	0:7c97dcef700c	1585	p += 5;
steiger	0:7c97dcef700c	1586
steiger	0:7c97dcef700c	1587	version = "";
steiger	0:7c97dcef700c	1588	encoding = "";
steiger	0:7c97dcef700c	1589	standalone = "";
steiger	0:7c97dcef700c	1590
steiger	0:7c97dcef700c	1591	while ( p && *p )
steiger	0:7c97dcef700c	1592	{
steiger	0:7c97dcef700c	1593	if ( *p == '>' )
steiger	0:7c97dcef700c	1594	{
steiger	0:7c97dcef700c	1595	++p;
steiger	0:7c97dcef700c	1596	return p;
steiger	0:7c97dcef700c	1597	}
steiger	0:7c97dcef700c	1598
steiger	0:7c97dcef700c	1599	p = SkipWhiteSpace( p, _encoding );
steiger	0:7c97dcef700c	1600	if ( StringEqual( p, "version", true, _encoding ) )
steiger	0:7c97dcef700c	1601	{
steiger	0:7c97dcef700c	1602	TiXmlAttribute attrib;
steiger	0:7c97dcef700c	1603	p = attrib.Parse( p, data, _encoding );
steiger	0:7c97dcef700c	1604	version = attrib.Value();
steiger	0:7c97dcef700c	1605	}
steiger	0:7c97dcef700c	1606	else if ( StringEqual( p, "encoding", true, _encoding ) )
steiger	0:7c97dcef700c	1607	{
steiger	0:7c97dcef700c	1608	TiXmlAttribute attrib;
steiger	0:7c97dcef700c	1609	p = attrib.Parse( p, data, _encoding );
steiger	0:7c97dcef700c	1610	encoding = attrib.Value();
steiger	0:7c97dcef700c	1611	}
steiger	0:7c97dcef700c	1612	else if ( StringEqual( p, "standalone", true, _encoding ) )
steiger	0:7c97dcef700c	1613	{
steiger	0:7c97dcef700c	1614	TiXmlAttribute attrib;
steiger	0:7c97dcef700c	1615	p = attrib.Parse( p, data, _encoding );
steiger	0:7c97dcef700c	1616	standalone = attrib.Value();
steiger	0:7c97dcef700c	1617	}
steiger	0:7c97dcef700c	1618	else
steiger	0:7c97dcef700c	1619	{
steiger	0:7c97dcef700c	1620	// Read over whatever it is.
steiger	0:7c97dcef700c	1621	while( p && p && p != '>' && !IsWhiteSpace( *p ) )
steiger	0:7c97dcef700c	1622	++p;
steiger	0:7c97dcef700c	1623	}
steiger	0:7c97dcef700c	1624	}
steiger	0:7c97dcef700c	1625	return 0;
steiger	0:7c97dcef700c	1626	}
steiger	0:7c97dcef700c	1627
steiger	0:7c97dcef700c	1628	bool TiXmlText::Blank() const
steiger	0:7c97dcef700c	1629	{
steiger	0:7c97dcef700c	1630	for ( unsigned i=0; i<value.length(); i++ )
steiger	0:7c97dcef700c	1631	if ( !IsWhiteSpace( value[i] ) )
steiger	0:7c97dcef700c	1632	return false;
steiger	0:7c97dcef700c	1633	return true;
steiger	0:7c97dcef700c	1634	}
steiger	0:7c97dcef700c	1635

Repository toolbox

Repository details

Type:	Library
Created:	21 Nov 2010
Imports:	11
Forks:	0
Commits:	1
Dependents:	0
Dependencies:	0
Followers:	4

Important changes to repositories hosted on mbed.com

tinyxmlparser.cpp@0:7c97dcef700c, 2010-11-21 (annotated)

Who changed what in which revision?

Repository toolbox

Repository details

Important Information for this Arm website

Important changes to repositories hosted on mbed.com

tinyxmlparser.cpp@0:7c97dcef700c, 2010-11-21 (annotated)

Who changed what in which revision?

Repository toolbox

Repository details

Important Information for this Arm website

Access Warning