public function extractPage($pageID, $pageTitle, $pageSource)
{
$result = new ExtractionResult($pageID, $this->language, self::extractorID);
$pageID = encodeLocalName($pageID);
// Remove Template as this is already extracted by the Infobox Extractor
// Find subtemplates and remove Subtemplates, which are listed as ignored!
preg_match_all('~\\{((?>[^{}]+)|(?R))*\\}~x', $pageSource, $subTemplates);
foreach ($subTemplates[0] as $key => $subTemplate) {
$subTemplate = preg_replace("/(^\\{\\{)|(\\}\\}\$)/", "", $subTemplate);
// Cut Brackets / {}
$pageSource = str_replace('{{' . $subTemplate . '}}', '', $pageSource);
}
// Extract internal Semantic Links
$findSemanticLinks = preg_match_all('/(\\[\\[)([a-zA-z0-9\\- _]+)(::)([^\\]]+)\\]\\]/', $pageSource, $matches, PREG_SET_ORDER);
foreach ($matches as $match) {
$result->addTriple(RDFtriple::page($pageID), RDFtriple::predicate(encodeLocalName($match[2])), RDFtriple::page($match[4]));
}
// Extract Literals
$findSemanticLinks = preg_match_all('/(\\[\\[)([a-zA-Z\\-_ ]+)(:=)([^\\]]+)\\]\\]/', $pageSource, $matches, PREG_SET_ORDER);
foreach ($matches as $match) {
$triple = array();
$triple = parseAttributeValue($match[4], $pageID, $match[2]);
// object, object_is, datatype(, language)
$lexicalForm = $triple[0];
$datatype = $triple[2];
$predicate = propertyToCamelCase(encodeLocalName($match[2]));
// Continue if empty String
if ($lexicalForm == null) {
continue;
}
$result->addTriple(RDFtriple::page($pageID), RDFTriple::predicate($predicate), RDFtriple::literal($lexicalForm, $datatype, 'en'));
}
return $result;
}
/**
* @param $shortDestination destination for short abstract.
* If null, no short abstracts will be extracted.
* @param $longDestination destination for long abstract.
* If null, no long abstracts will be extracted.
*/
public function __construct()
{
parent::__construct();
//no validation required
$this->shortPredicate = RDFtriple::URI(RDFS_COMMENT, false);
$this->longPredicate = RDFtriple::predicate("abstract");
}
public function extractPage($pageID, $pageTitle, $pageSource)
{
$result = new ExtractionResult($pageID, $this->language, self::extractorID);
$CleanSource = $this->remove_wikicode($pageSource);
$LongAbstract = $this->extract_abstract($CleanSource, 3000, false, false);
$result->addTriple(RDFtriple::page($pageID), RDFtriple::predicate("abstract"), RDFtriple::Literal($LongAbstract, NULL, $this->language));
return $result;
}
public function extractPage($pageID, $pageTitle, $pageSource)
{
// language code in URI uses '-', not '_'
$language = str_replace('_', '-', $this->language);
$subject = $this->getPageURI();
$predicate = $language == 'en' ? RDFtriple::predicate("wikipage-" . $language) : RDFtriple::URI(FOAF_PAGE, false);
$object = RDFtriple::URI("http://" . $language . ".wikipedia.org/wiki/" . URI::wikipediaEncode($pageTitle));
$result = new ExtractionResult($pageID, $this->language, $this->getExtractorID());
$result->addTriple($subject, $predicate, $object);
return $result;
}
public function extractPage($pageID, $pageTitle, $pageSource)
{
$result = new ExtractionResult($pageID, $this->language, $this->getExtractorID());
if ($this->language == "en") {
$query = "select wsl.url from templatelinks tl inner join page p on p.page_id = tl.tl_from\r\n\t\t\t\t\t\tinner join dbpedia_develop.wordnet_mapping wm on tl.tl_title = wm.infobox\r\n\t\t\t\t\t\tinner join dbpedia_develop.wordnet_synsets_links wsl on wm.ID1 = wsl.synset30ID\r\n\t\t\t\t\t\twhere p.page_title = '" . mysql_escape_string($pageID) . "' and p.page_namespace = 0";
$queryresult = mysql_query($query, $this->DBlink) or die("Query failed:\n{$query}\n" . mysql_error());
while ($row = mysql_fetch_array($queryresult, MYSQL_ASSOC)) {
$result->addTriple(RDFtriple::page($pageID), RDFtriple::predicate("wordnet_type"), RDFtriple::URI($row["url"]));
}
}
return $result;
}
public function extractPage($pageID, $pageTitle, $pageSource)
{
$result = new ExtractionResult($pageID, $this->language, self::extractorID);
$PersonData = $this->extractPersondata($pageSource, $this->language);
//var_dump($PersonData);
if ($PersonData != null) {
// preg_match("/\[\[en:(.*)\]\]/", $pageSource, $LangLinkmatch);
// $PersonData['enPageID'] = str_replace(" ","_",$LangLinkmatch[1]);
$WikiDB = new DatabaseWikipedia($this->language);
preg_match("/\\[\\[([^\\]]*)\\]\\]/", $PersonData['birthplace'], $Birthplacematch);
$Birthplacematch = $this->getLinkForLabeledLink($Birthplacematch);
$mySource = $WikiDB->getSource($Birthplacematch);
preg_match("/\\[\\[en:(.*)\\]\\]/", $mySource, $LangLinkmatch);
$BirthPlace = $LangLinkmatch[1];
preg_match("/\\[\\[([^\\]]*)\\]\\]/", $PersonData['deathplace'], $Deathplacematch);
$Deathplacematch = $this->getLinkForLabeledLink($Deathplacematch);
$mySource = $WikiDB->getSource($Deathplacematch);
preg_match("/\\[\\[en:(.*)\\]\\]/", $mySource, $LangLinkmatch);
$DeathPlace = $LangLinkmatch[1];
//var_dump($PersonData);
//var_dump($BirthPlace);
//var_dump($DeathPlace);
//var_dump($Deathplacematch);
$result->addTriple(RDFtriple::page($pageID), RDFtriple::URI("http://xmlns.com/foaf/0.1/name"), RDFtriple::Literal($PersonData['name'], null, "de"));
$result->addTriple(RDFtriple::page($pageID), RDFtriple::URI("http://xmlns.com/foaf/0.1/givenname"), RDFtriple::Literal($PersonData['givenname'], null, "de"));
$result->addTriple(RDFtriple::page($pageID), RDFtriple::URI("http://xmlns.com/foaf/0.1/surname"), RDFtriple::Literal($PersonData['surname'], null, "de"));
if ($BirthPlace != "") {
$result->addTriple(RDFtriple::page($pageID), RDFtriple::predicate("birthPlace"), RDFtriple::page($BirthPlace));
// $result->addTriple(
// RDFtriple::page($pageID),
// RDFtriple::URI("http://purl.org/vocab/bio/0.1/event"),
// RDFtriple::URI("http://dbpedia.org/resource/" . URI::wikipediaEncode($pageID) . "/Birth"));
// $result->addTriple(
// RDFtriple::URI("http://dbpedia.org/resource/" . URI::wikipediaEncode($pageID) . "/Birth");
// RDFtriple::URI("http://purl.org/vocab/bio/0.1/place"),
// RDFtriple::page($BirthPlace));
}
if ($PersonData['birthdate'] != "") {
$result->addTriple(RDFtriple::page($pageID), RDFtriple::predicate("birth"), RDFtriple::Literal($PersonData['birthdate'], "http://www.w3.org/2001/XMLSchema#date", null));
}
if ($DeathPlace != "") {
$result->addTriple(RDFtriple::page($pageID), RDFtriple::predicate("deathPlace"), RDFtriple::page($DeathPlace));
}
if ($PersonData['deathdate'] != "") {
$result->addTriple(RDFtriple::page($pageID), RDFtriple::predicate("death"), RDFtriple::Literal($PersonData['deathdate'], "http://www.w3.org/2001/XMLSchema#date", null));
}
if ($PersonData['description'] != "") {
$result->addTriple(RDFtriple::page($pageID), RDFtriple::URI("http://purl.org/dc/elements/1.1/description"), RDFtriple::Literal($PersonData['description'], null, "de"));
}
$result->addTriple(RDFtriple::page($pageID), RDFtriple::URI("http://www.w3.org/1999/02/22-rdf-syntax-ns#type"), RDFtriple::URI("http://xmlns.com/foaf/0.1/Person"));
}
return $result;
}
public function extractPage($pageID, $pageTitle, $pageSource)
{
$result = new ExtractionResult($pageID, $this->language, self::extractorID);
$extlinks = $this->extract_external_links($pageSource, $this->language);
while (list($ExtURL, $ExtName) = each($extlinks)) {
if (!URI::validate($ExtURL)) {
continue;
}
$result->addTriple(RDFtriple::page($pageID), RDFtriple::predicate("reference"), RDFtriple::URI($ExtURL));
}
return $result;
}
public function extractPage($pageID, $pageTitle, $pageSource)
{
//create a new Extraction Result to hold all extrated Triples
$result = new ExtractionResult($pageID, $this->language, self::extractorID);
//Look for {{chembox header}} in PageSource
if (preg_match("/{{chembox header}}/", $pageSource, $match)) {
//DO SOME PARSING
//Add a Triple for each Property
$result->addTriple(RDFtriple::page($pageID), RDFtriple::predicate("my_chem_property"), RDFtriple::Literal("my_value"));
//Add each Predicate to the Predicate Collection
$this->allPredicates->addPredicate("my_chem_property");
}
return $result;
}
public function extractPage($pageID, $pageTitle, $pageSource)
{
$result = new ExtractionResult($pageID, $this->language, $this->getExtractorID());
if (Util::isRedirect($pageSource, $this->language)) {
if (preg_match_all("/\\[\\[([^\\]]*)\\]\\]/", $pageSource, $matches, PREG_SET_ORDER)) {
foreach ($matches as $match) {
$s = RDFtriple::page($pageID);
$p = RDFtriple::predicate("redirect");
$o = RDFtriple::page($this->getLinkForLabeledLink($match[1]));
$templateredirecturi = str_replace("template:", "Template:", str_replace("'", "\\'", str_replace(" ", "_", mb_strtolower($match[1]))));
$templateredirecturi = DB_RESOURCE_NS . $templateredirecturi;
$query = "select * from template_uri where uri = '{$templateredirecturi}'";
$dbresult = mysql_query($query, $this->link) or die("Query failed: " . mysql_error() . ' - ' . $query);
$uri = "";
while ($row = mysql_fetch_array($dbresult, MYSQL_ASSOC)) {
$uri = $row['uri'];
$template_id = $row['template_id'];
/*
echo "$this->redirectTemplateCounter: $pageID => $match[1]";
echo " $uri (FOUND)";
echo "\n";
*/
$newtemplateuri = DB_RESOURCE_NS . str_replace("template:", "Template:", str_replace("'", "\\'", str_replace(" ", "_", mb_strtolower($pageID))));
$query_template_uri = "select * from template_uri where uri = '{$newtemplateuri}'";
$dbresult_template_uri = mysql_query($query_template_uri, $this->link) or die("Query failed: " . mysql_error() . ' - ' . $query_template_uri);
if (mysql_num_rows($dbresult_template_uri) > 0) {
//echo "$pageID already in DB";
} else {
$insertquery = "INSERT INTO template_uri (template_id, uri) VALUES ('" . $template_id . "', '" . $newtemplateuri . "')";
mysql_query($insertquery, $this->link) or die("Query failed: " . mysql_error() . ' - ' . $insertquery);
$this->redirectTemplateCounter++;
echo "{$this->redirectTemplateCounter}: {$pageID} => {$match['1']}";
echo " {$newtemplateuri} (FOUND)";
echo "\n";
}
}
}
}
}
return $result;
}
public function extractPage($pageID, $pageTitle, $pageSource)
{
$result = new ExtractionResult($pageID, $this->language, self::extractorID);
$result->addTriple(RDFtriple::page($pageID), RDFtriple::predicate("characterCount"), RDFtriple::literal(strlen($pageSource)));
return $result;
}
