source: gaycko/trunk/src/parse.c Tweet

/*
 *  Gaycko        Text mode web browser
 *  Copyright (c) 2011 Jean-Yves Lamoureux <jylam@lnxscene.org>
 *                All Rights Reserved
 *
 *  This library is free software. It comes without any warranty, to
 *  the extent permitted by applicable law. You can redistribute it
 *  and/or modify it under the terms of the Do What The Fuck You Want
 *  To Public License, Version 2, as published by Sam Hocevar. See
 *  http://sam.zoy.org/wtfpl/COPYING for more details.
 */
#include "parse.h"

gDOM *gaycko_parse(char *data, unsigned int size) {

        /* Clean up and repair bad HTML */
        TidyDoc tdoc = tidyCreate();
        TidyBuffer errbuf = {0};

        TidyBuffer output = {0};

        Bool ok;
        int rc = -1;

        ok = tidyOptSetBool( tdoc, TidyXhtmlOut, yes );  // Convert to XHTML
        if ( ok )
                rc = tidySetErrorBuffer( tdoc, &errbuf );      // Capture diagnostics
        if ( rc >= 0 )
                rc = tidyParseString( tdoc, data );           // Parse the input
        if ( rc >= 0 )
                rc = tidyCleanAndRepair( tdoc );               // Tidy it up!
        if ( rc > 1 )                                    // If error, force output.
                rc = ( tidyOptSetBool(tdoc, TidyForceOutput, yes) ? rc : -1 );
        if ( rc >= 0 )
                rc = tidySaveBuffer( tdoc, &output );          // Pretty Print

        /* Actual parsing */
        htmlDocPtr doc = htmlParseDoc((unsigned char*)output.bp, NULL);
        
        /* Release tidy document */
        tidyRelease( tdoc );

        /* Convert libxml2's tree to our own DOM */
        gDOM *dom = gaycko_convert_dom(doc);

        /* Free libxml2 tree */
        xmlFreeDoc(doc);

        return dom;
}