diff options
Diffstat (limited to 'src/contrib/xml/tinyxmlparser.cpp')
| -rw-r--r-- | src/contrib/xml/tinyxmlparser.cpp | 2684 |
1 files changed, 1372 insertions, 1312 deletions
diff --git a/src/contrib/xml/tinyxmlparser.cpp b/src/contrib/xml/tinyxmlparser.cpp index f06d1cd..9c696b6 100644 --- a/src/contrib/xml/tinyxmlparser.cpp +++ b/src/contrib/xml/tinyxmlparser.cpp @@ -1,23 +1,23 @@ /* 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. */ @@ -29,1465 +29,1525 @@ distribution. // 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 ] = -{ - { "&", 5, '&' }, - { "<", 4, '<' }, - { ">", 4, '>' }, - { """, 6, '\"' }, - { "'", 6, '\'' } -}; +// or order will break putstring. +TiXmlBase::Entity TiXmlBase::entity[ NUM_ENTITY ] = + { + { "&", 5, '&' }, + { "<", 4, '<' }, + { ">", 4, '>' }, + { """, 6, '\"' }, + { "'", 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: +// Beware of the non-characters in UTF-8: // ef bb bf (Microsoft "lead bytes") // ef bf be -// ef bf bf - - - -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 -}; +// ef bf bf + + + +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]); - } + 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 ); -// } + // 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 ); -// } + // 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; + 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 ) - { - // Code contributed by Fletcher Dunn: (modified by lee) - switch (*p) { - 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 (char)(0xef): - if ( encoding == TIXML_ENCODING_UTF8 ) - { - if ( *(p+1) && *(p+2) ) - { - // In these cases, don't advance the column. These are - // 0-width spaces. - if ( *(p+1)==(char)(0xbb) && *(p+2)==(char)(0xbf) ) - p += 3; - else if ( *(p+1)==(char)(0xbf) && *(p+2)==(char)(0xbe) ) - p += 3; - else if ( *(p+1)==(char)(0xbf) && *(p+2)==(char)(0xbf) ) - 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 ); + 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 ) + { + // Code contributed by Fletcher Dunn: (modified by lee) + switch (*p) + { + 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 (char)(0xef): + if ( encoding == TIXML_ENCODING_UTF8 ) + { + if ( *(p+1) && *(p+2) ) + { + // In these cases, don't advance the column. These are + // 0-width spaces. + if ( *(p+1)==(char)(0xbb) && *(p+2)==(char)(0xbf) ) + p += 3; + else if ( *(p+1)==(char)(0xbf) && *(p+2)==(char)(0xbe) ) + p += 3; + else if ( *(p+1)==(char)(0xbf) && *(p+2)==(char)(0xbf) ) + 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 ) - { - // Skip the stupid Microsoft UTF-8 Byte order marks - if ( *(p+0)==(char) 0xef - && *(p+1)==(char) 0xbb - && *(p+2)==(char) 0xbf ) - { - p += 3; - continue; - } - else if(*(p+0)==(char) 0xef - && *(p+1)==(char) 0xbf - && *(p+2)==(char) 0xbe ) - { - p += 3; - continue; - } - else if(*(p+0)==(char) 0xef - && *(p+1)==(char) 0xbf - && *(p+2)==(char) 0xbf ) - { - 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; + if ( !p || !*p ) + { + return 0; + } + if ( encoding == TIXML_ENCODING_UTF8 ) + { + while ( *p ) + { + // Skip the stupid Microsoft UTF-8 Byte order marks + if ( *(p+0)==(char) 0xef + && *(p+1)==(char) 0xbb + && *(p+2)==(char) 0xbf ) + { + p += 3; + continue; + } + else if(*(p+0)==(char) 0xef + && *(p+1)==(char) 0xbf + && *(p+2)==(char) 0xbe ) + { + p += 3; + continue; + } + else if(*(p+0)==(char) 0xef + && *(p+1)==(char) 0xbf + && *(p+2)==(char) 0xbf ) + { + 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(); - } + 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; + //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; + *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; - unsigned 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; + // 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; + unsigned 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 ) + const char* tag, + bool ignoreCase, + TiXmlEncoding encoding ) { - assert( p ); - assert( tag ); - if ( !p || !*p ) - { - assert( 0 ); - return false; - } - - const char* q = p; - - if ( ignoreCase ) - { - while ( *q && *tag && ToLower( *q, encoding ) == ToLower( *tag, encoding ) ) - { - ++q; - ++tag; - } - - if ( *tag == 0 ) - return true; - } - else - { - while ( *q && *tag && *q == *tag ) - { - ++q; - ++tag; - } - - if ( *tag == 0 ) // Have we found the end of the tag, and everything equal? - return true; - } - return false; + assert( p ); + assert( tag ); + if ( !p || !*p ) + { + assert( 0 ); + return false; + } + + const char* q = p; + + if ( ignoreCase ) + { + while ( *q && *tag && ToLower( *q, encoding ) == ToLower( *tag, encoding ) ) + { + ++q; + ++tag; + } + + if ( *tag == 0 ) + return true; + } + else + { + while ( *q && *tag && *q == *tag ) + { + ++q; + ++tag; + } + + if ( *tag == 0 ) // Have we found the end of the tag, and everything equal? + return true; + } + return false; } -const char* TiXmlBase::ReadText( const char* p, - TIXML_STRING * text, - bool trimWhiteSpace, - const char* endTag, - bool caseInsensitive, - TiXmlEncoding encoding ) +const char* TiXmlBase::ReadText( const char* p, + TIXML_STRING * text, + bool trimWhiteSpace, + const char* endTag, + bool caseInsensitive, + TiXmlEncoding encoding ) { - *text = ""; - if ( !trimWhiteSpace // certain tags always keep whitespace - || !condenseWhiteSpace ) // if true, whitespace is always kept - { - // Keep all the white space. - while ( p && *p - && !StringEqual( p, endTag, caseInsensitive, encoding ) - ) - { - int len; - char cArr[4] = { 0, 0, 0, 0 }; - p = GetChar( p, cArr, &len, encoding ); - text->append( cArr, len ); - } - } - else - { - bool whitespace = false; - - // Remove leading white space: - p = SkipWhiteSpace( p, encoding ); - while ( p && *p - && !StringEqual( p, endTag, caseInsensitive, encoding ) ) - { - if ( *p == '\r' || *p == '\n' ) - { - whitespace = true; - ++p; - } - else if ( IsWhiteSpace( *p ) ) - { - whitespace = true; - ++p; - } - else - { - // If we've found whitespace, add it before the - // new character. Any whitespace just becomes a space. - if ( whitespace ) - { - (*text) += ' '; - whitespace = false; - } - int len; - char cArr[4] = { 0, 0, 0, 0 }; - p = GetChar( p, cArr, &len, encoding ); - if ( len == 1 ) - (*text) += cArr[0]; // more efficient - else - text->append( cArr, len ); - } - } - } - return p + strlen( endTag ); + *text = ""; + if ( !trimWhiteSpace // certain tags always keep whitespace + || !condenseWhiteSpace ) // if true, whitespace is always kept + { + // Keep all the white space. + while ( p && *p + && !StringEqual( p, endTag, caseInsensitive, encoding ) + ) + { + int len; + char cArr[4] = { 0, 0, 0, 0 }; + p = GetChar( p, cArr, &len, encoding ); + text->append( cArr, len ); + } + } + else + { + bool whitespace = false; + + // Remove leading white space: + p = SkipWhiteSpace( p, encoding ); + while ( p && *p + && !StringEqual( p, endTag, caseInsensitive, encoding ) ) + { + if ( *p == '\r' || *p == '\n' ) + { + whitespace = true; + ++p; + } + else if ( IsWhiteSpace( *p ) ) + { + whitespace = true; + ++p; + } + else + { + // If we've found whitespace, add it before the + // new character. Any whitespace just becomes a space. + if ( whitespace ) + { + (*text) += ' '; + whitespace = false; + } + int len; + char cArr[4] = { 0, 0, 0, 0 }; + p = GetChar( p, cArr, &len, encoding ); + if ( len == 1 ) + (*text) += cArr[0]; // more efficient + else + text->append( cArr, len ); + } + } + } + return p + strlen( endTag ); } #ifdef TIXML_USE_STL void TiXmlDocument::StreamIn( TIXML_ISTREAM * in, TIXML_STRING * tag ) { - // The basic issue with a document is that we don't know what we're - // streaming. Read something presumed to be a tag (and hope), then - // identify it, and call the appropriate stream method on the tag. - // - // This "pre-streaming" will never read the closing ">" so the - // sub-tag can orient itself. - - if ( !StreamTo( in, '<', tag ) ) - { - SetError( TIXML_ERROR_PARSING_EMPTY, 0, 0, TIXML_ENCODING_UNKNOWN ); - return; - } - - while ( in->good() ) - { - int tagIndex = (int) tag->length(); - while ( in->good() && in->peek() != '>' ) - { - int c = in->get(); - if ( c <= 0 ) - { - SetError( TIXML_ERROR_EMBEDDED_NULL, 0, 0, TIXML_ENCODING_UNKNOWN ); - break; - } - (*tag) += (char) c; - } - - if ( in->good() ) - { - // We now have something we presume to be a node of - // some sort. Identify it, and call the node to - // continue streaming. - TiXmlNode* node = Identify( tag->c_str() + tagIndex, TIXML_DEFAULT_ENCODING ); - - if ( node ) - { - node->StreamIn( in, tag ); - bool isElement = node->ToElement() != 0; - delete node; - node = 0; - - // If this is the root element, we're done. Parsing will be - // done by the >> operator. - if ( isElement ) - { - return; - } - } - else - { - SetError( TIXML_ERROR, 0, 0, TIXML_ENCODING_UNKNOWN ); - return; - } - } - } - // We should have returned sooner. - SetError( TIXML_ERROR, 0, 0, TIXML_ENCODING_UNKNOWN ); + // The basic issue with a document is that we don't know what we're + // streaming. Read something presumed to be a tag (and hope), then + // identify it, and call the appropriate stream method on the tag. + // + // This "pre-streaming" will never read the closing ">" so the + // sub-tag can orient itself. + + if ( !StreamTo( in, '<', tag ) ) + { + SetError( TIXML_ERROR_PARSING_EMPTY, 0, 0, TIXML_ENCODING_UNKNOWN ); + return; + } + + while ( in->good() ) + { + int tagIndex = (int) tag->length(); + while ( in->good() && in->peek() != '>' ) + { + int c = in->get + (); + if ( c <= 0 ) + { + SetError( TIXML_ERROR_EMBEDDED_NULL, 0, 0, TIXML_ENCODING_UNKNOWN ); + break; + } + (*tag) += (char) c; + } + + if ( in->good() ) + { + // We now have something we presume to be a node of + // some sort. Identify it, and call the node to + // continue streaming. + TiXmlNode* node = Identify( tag->c_str() + tagIndex, TIXML_DEFAULT_ENCODING ); + + if ( node ) + { + node->StreamIn( in, tag ); + bool isElement = node->ToElement() != 0; + delete node; + node = 0; + + // If this is the root element, we're done. Parsing will be + // done by the >> operator. + if ( isElement ) + { + return; + } + } + else + { + SetError( TIXML_ERROR, 0, 0, TIXML_ENCODING_UNKNOWN ); + return; + } + } + } + // We should have returned sooner. + SetError( TIXML_ERROR, 0, 0, TIXML_ENCODING_UNKNOWN ); } #endif const char* TiXmlDocument::Parse( const char* p, TiXmlParsingData* prevData, TiXmlEncoding encoding ) { - ClearError(); - - // Parse away, at the document level. Since a document - // contains nothing but other tags, most of what happens - // here is skipping white space. |