gistfile1.php

<?php

if (!$sContent = file_get_contents('http://www.biblija.net/biblija.cgi?m=lc1,1-10&id42=0&l=nl'))
    throw new Exception('Tekst kon niet geladen worden');

    // Try to get encoding right, lose the strange encoding chars
$sContent = mb_convert_encoding($sContent, 'UTF-8', mb_detect_encoding($sContent));
$sContent = mb_convert_encoding($sContent, 'html-entities', 'UTF-8');

$oDOM = new DOMDocument('1.0');
//$oDOM->resolveExternals = false;
//$oDOM->substituteEntities = false; 

if (!$oDOM->loadHTML($sContent))
    throw new Exception('Kon tekst niet parsen');

//if (strstr($oDOM->documentElement->textContent, 'Â'))
//    throw new Exception('Get the bastard!');
    
    // Let op, alle tags en attributes zijn naar lowercase omgezet.
if (!$oQry = new DOMXPath($oDOM))
    throw new Exception('Could not parse XPath query');

$oTexts = $oQry->query('//td[@class = "text"]');

    // The element containing the content
foreach ($oTexts  as $oText)
{   
    $oParas = $oQry->query('div', $oText);

        // Loop through all parts found
    foreach ($oParas as $oPara)
    {   
            // What sort of chapter we have?
        switch ($oPara->getAttribute('class'))
        {
            case 'm':
            case 'p':
                // Doorloop childs, check nodeType (text, element > span? met class 'v') of doe een regexp-setje ;)
                
                // \v = vertical whitespace
                $aVerses = preg_split('/\v/is', $oPara->textContent);                    
                
                foreach ($aVerses as $sVerse)
                {   
                    if (trim($sVerse) != '')
                    {
                        // \h = horizontal whitespace, later limit 2
                        $aResult = preg_split('/\h+/is', $sVerse, 2);
                        
                            // Fix strange encoding bug
                        foreach ($aResult as $iIndex => $sResult) $aResult[$iIndex] = preg_replace('/Â/','',$sResult);
                        
                        print_r($aResult);
                    }
                }
            break;
            
            default:
               //throw new Exception('Paragraph-type `'.$oPara->getAttribute('class').'` is not handled.'); 
        }
    }
}

?>