cppAdaptive1

 view release on metacpan or  search on metacpan

src/dlib/xml_parser/xml_parser_kernel_1.h  view on Meta::CPAN

            
            // -----------------------------------

            enum token_type
            {
                element_start, // the first tag of an element
                element_end,   // the last tag of an element
                empty_element, // the singular tag of an empty element
                pi,            // processing instruction
                chars,         // the non-markup data between tags
                chars_cdata,   // the data from a CDATA section
                eof,           // this token is returned when we reach the end of input
                error,         // this token indicates that the tokenizer couldn't
                               // determine which category the next token fits into
                dtd,           // this token is for an entire dtd 
                comment        // this is a token for comments
            };
            /*
                notes about the tokens:
                    the tokenizer guarantees that the following tokens to not 
                    contain the '<' character except as the first character of the token

src/dlib/xml_parser/xml_parser_kernel_1.h  view on Meta::CPAN

            // notify all the document_handlers that we are about to being parsing
            for (unsigned long i = 0; i < dh_list.size(); ++i)
            {
                dh_list[i]->start_document();
            }


            std::string chars_buf; // used to collect chars data between consecutive
                                // chars and chars_cdata tokens so that 
                                // document_handlers receive all chars data between
                                // tags in one call

            // variables to be used with the parsing functions
            attrib_list atts;
            std::string name;
            std::string target;
            std::string data;

            

src/dlib/xml_parser/xml_parser_kernel_1.h  view on Meta::CPAN

                        }
                        else if (token_text.find_first_not_of(" \t\r\n") != std::string::npos)
                        {
                            // you can't have non whitespace chars data outside the root element
                            seen_fatal_error = true;                        
                        }
                    }break;

                // ----------------------------------------

                case chars_cdata:
                    {
                        if (tags.size() != 0)
                        {
                            chars_buf += token_text;
                        }
                        else
                        {
                            // you can't have chars_data outside the root element
                            seen_fatal_error = true;
                        }

src/dlib/xml_parser/xml_parser_kernel_1.h  view on Meta::CPAN

                if (seen_fatal_error)
                    break;

                // if we have seen the last tag then quit the loop
                if (tags.size() == 0 && seen_root_tag)
                    break;
                

                get_next_token(in,token_text,token_kind,line_number);

                // if the next token is not a chars or chars_cdata token then flush
                // the chars_buf to the document_handlers
                if ( (token_kind != chars) && 
                    (token_kind != chars_cdata) &&
                    (token_kind != dtd) && 
                    (token_kind != comment) &&
                    (chars_buf.size() != 0)
                    )
                {
                    // notify all the document_handlers
                    for (unsigned long i = 0; i < dh_list.size(); ++i)
                    {
                        dh_list[i]->characters(chars_buf);
                    }

src/dlib/xml_parser/xml_parser_kernel_1.h  view on Meta::CPAN

            // this is the start of some kind of a tag
        case '<':
            {
                ch2 = in.get();
                switch (ch2)
                {
                
                // ---------------------------------

                    // this is a dtd, comment, or chars_cdata token 
                case '!':
                    {
                        // if this is a CDATA section *******************************
                        if ( in.peek() == '[')
                        {
                            token_kind = chars_cdata;

                            // throw away the '['
                            in.get();

                            // make sure the next chars are CDATA[
                            std::istream::int_type ch = in.get();
                            if (ch != 'C')                                
                                token_kind = error;
                            ch = in.get();
                            if (ch != 'D')