HTML_Tree-2.4.3.rar_HTML-Tree-4.2.tar.gz_html tree_install html

/* ** HTML Tree ** html.c ** ** Copyright (C) 1999 Paul J. Lucas ** ** This program is free software; you can redistribute it and/or modify ** it under the terms of the GNU General Public License as published by ** the Free Software Foundation; either version 2 of the License, or ** (at your option) any later version. ** ** This program is distributed in the hope that it will be useful, ** but WITHOUT ANY WARRANTY; without even the implied warranty of ** MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the ** GNU General Public License for more details. ** ** You should have received a copy of the GNU General Public License ** along with this program; if not, write to the Free Software ** Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ // standard #include <cctype> #include <cstring> #include <string> // local #include "Comment_Node.h" #include "Content_Node.h" #include "Element_Node.h" #include "fake_ansi.h" #include "html.h" #include "Text_Node.h" #include "util.h" #ifndef PJL_NO_NAMESPACES using namespace std; namespace HTML_Tree { #endif int const Tag_Name_Max_Size = 10; // The maximum size of an HTML tag name, e.g., "BLOCKQUOTE". You // might need to increase this if you are indexing HTML documents // that contain non-standard tags and at least one of them is // longer than the above. void parse_html_comment( char const *&pos, char const *end ); void parse_html_tag( char const *&pos, char const *end, Content_Node*& ); void skip_html_tag( char const *&pos, char const *end ); //***************************************************************************** // // SYNOPSIS // Root_Node* html_parse( register char const *c, char const *end, bool include_comments ) // // DESCRIPTION // // Parse the HTML file to build a tree. // // PARAMETERS // // c The pointer to the beginning of the buffer. // // end The pointer to one past the end of the buffer. // // RETURN VALUE // // Returns a pointer to the root node of the tree. // //***************************************************************************** { Root_Node *const root_node = new Root_Node; Content_Node *cur_node = root_node; while ( c != end ) { char const *const b = c; if ( *c++ == '<' ) { if ( c == end ) break; if ( *c != '!' ) { // // It must be an HTML tag. // parse_html_tag( c, end, cur_node ); continue; } if ( ++c == end ) break; if ( *c == '-' && c + 1 != end && c[1] == '-' ) { // // It's a comment. // parse_html_comment( c += 2, end ); if ( include_comments ) new Comment_Node( b, c, cur_node ); continue; } // // This is something like <!DOCUMENT // skip_html_tag( c, end ); goto new_text_node; } // // Collect a run of text into a Text_Node. // for ( ; c != end; ++c ) if ( *c == '<' ) { // // We've encountered the potential beginning of // an HTML tag: stop collecting text and create // a new Text_Node containing what we've // collected so far. // break; } new_text_node: new Text_Node( b, c, cur_node ); } if ( root_node->empty() ) { delete root_node; return 0; } return root_node; } //***************************************************************************** // // SYNOPSIS // void parse_attributes( register char const *c, char const *end, Element_Node::attribute_map &attributes ) // // DESCRIPTION // // Parse out all the attributes and their values of an HTML element. // // PARAMETERS // // c The iterator marking the beginning of where to start // parsing. // // end The iterator marking the end of where to stop parsing // (usually positioned at the closing '>' character of the // HTML tag). // // attributes Where to deposit the attributes. // // SEE ALSO // // Dave Raggett, Arnaud Le Hors, and Ian Jacobs. "On SGML and HTML: SGML // constructs used in HTML: Attributes," HTML 4.0 Specification, section // 3.2.2, World Wide Web Consortium, April 1998. // http://www.w3.org/TR/PR-html40/intro/sgmltut.html#h-3.2.2 // //***************************************************************************** { while ( c != end && !is_space( *c++ ) ) ; // skip element name while ( c != end ) { if ( !isalpha( *c ) ) { ++c; continue; } // // We just found the start of a potential attribute name: now // try to find its end by stopping at one of 'end', whitespace, // or '='. To be robust, we allow whitespace around the '='. // char const *const name_begin = c; while ( c != end && !isspace( *c ) && *c != '=' ) ++c; char const *const name_end = c; while ( c != end && isspace( *c ) ) ++c; string const name( to_lower( name_begin, name_end ) ); if ( c == end || *c != '=' ) { // // It's a Boolean attribute: set its value to be its // own name (per the HTML 4.0 spec). // attributes[ name ] = name; continue; } if ( c == end ) break; // Here, 'c' is positioned at the '='. while ( ++c != end && isspace( *c ) ) ; if ( c == end ) break; // // Determine the span of the attribute's value: if it started // with a quote, it's terminated only by the matching closing // quote; if not, it's terminated by a whitespace character (or // running into 'end'). // // This is more lenient than the HTML 4.0 specification in that // it allows non-quoted values to contain characters other than // the set [A-Za-z0-9.-], i.e., any non-whitespace character. // char const quote = ( *c == '"' || *c == '\'' ) ? *c : 0; if ( quote && ++c == end ) break; char const *const b = c; for ( ; c != end; ++c ) if ( quote ) { // stop at matching quote only if ( *c == quote ) break; } else if ( isspace( *c ) ) break; // stop at whitespace attributes[ name ] = string( b, c ); if ( c == end ) break; ++c; } } //***************************************************************************** // // SYNOPSIS // void parse_html_comment( register char const *&c, char const *end ) // // DESCRIPTION // // Skip past an HTML comment by scanning for the closing "-->" character // sequence. Unlike skipping an ordinary HTML tag, quotes are not // significant and no attempt must be made either to "balance" them or to // ignore what is in between them. The HTML specification permits // whitespace between the "--" and the ">" (for some strange reason). // // This function is more lenient than the HTML 4.0 specification in that // it allows for a string of hyphens within a comment since this is common // in practice; the specification considers this to be an error. // // PARAMETERS // // c The iterator to use. It is presumed to start anywhere after // the initial "<!--"; it is left positioned after the "-->". // // end The pointer to the end of the file. // // SEE ALSO // // Dave Raggett, Arnaud Le Hors, and Ian Jacobs. "On SGML and HTML: SGML // constructs used in HTML: Comments," HTML 4.0 Specification, section // 3.2.4, World Wide Web Consortium, April 1998. // http://www.w3.org/TR/PR-html40/intro/sgmltut.html#h-3.2.4 // //***************************************************************************** { while ( c != end ) if ( *c++ == '-' && c != end && *c == '-' ) { char const *const d = c; while ( ++c != end && is_space( *c ) ) ; if ( c == end || *c++ == '>' ) break; c = d; } } //***************************************************************************** // // SYNOPSIS // void parse_html_tag( register char const *&c, register char const *end, Content_Node *&cur_node ) // // DESCRIPTION // // This function does everything skip_html_tag() does but additionally // builds a DOM-like (Document Object Model) tree of nodes. It does this // by knowing when to end elements. // // PARAMETERS // // c The iterator to use. It must be positioned at the // character after the '<'; it is repositioned at the // first character after the '>'. // // end The iterator marking the end of the file. //