private function parseOptionKeyValuePair($optionKeyValuePair)
{
$keyValue = explode('=', $optionKeyValuePair);
$key = trim($keyValue[0]);
$value = trim($keyValue[1]);
switch ($key) {
case 'vextwarning':
$this->options->setVendorExtensionIssuesAsWarnings(filter_var($value, FILTER_VALIDATE_BOOLEAN));
break;
case 'output':
$this->options->setOutputFormat($value);
break;
case 'lang':
$this->options->setLanguage($value);
break;
case 'warning':
$warningLevel = filter_var($value, FILTER_VALIDATE_INT, array('options' => array('min_range' => 0, 'default' => 2)));
$this->options->setWarningLevel($warningLevel);
break;
case 'medium':
$this->options->setMedium($value);
break;
case 'profile':
$this->options->setProfile($value);
break;
}
}
<?php
/**
* This file starts the DBpedia extraction process for abstracts.
*
* Warning: The script needs several days to complete on an average PC.
*/
error_reporting(E_ALL);
// automatically loads required classes
require 'dbpedia.php';
// set $extractionDir and $extractionLanguages
require 'extractionconfig.php';
$manager = new ExtractionManager();
// loop over all languages
foreach ($extractionLanguages as $currLanguage) {
Options::setLanguage($currLanguage);
$pageTitles = new ArticlesSqlIterator($currLanguage);
$job = new ExtractionJob(new DatabaseWikipediaCollection($currLanguage), $pageTitles);
$extractionDirLang = $extractionDir . '/' . $currLanguage . '/';
if (!is_dir($extractionDirLang)) {
mkdir($extractionDirLang);
}
// AbstractExtractor has references to its two destinations, see below
$group = new ExtractionGroup(new NullDestination());
$shortDestination = new csvNTripleDestination($extractionDirLang . "shortabstract_" . $currLanguage);
$longDestination = new csvNTripleDestination($extractionDirLang . "longabstract_" . $currLanguage);
$extractorInstance = new AbstractExtractor();
$extractorInstance->setDestinations($shortDestination, $longDestination);
$group->addExtractor($extractorInstance);
$job->addExtractionGroup($group);
$date = date(DATE_RFC822);
foreach ($it as $key => $metainfo) {
Timer::start("main::processing");
//****PREPROCESSING*****
//print_r($metainfo);
$pageTitle = $metainfo['pageTitle'];
$pageTitles = new ArrayObject(array($pageTitle));
$pageURI = '';
try {
$pageURI = RDFtriple::page($pageTitle);
} catch (Exception $e) {
Logger::warn('main: invalid uri for ' . $pageTitle);
continue;
}
Logger::info("Title: {$pageTitle} " . mb_detect_encoding($pageTitle) . "");
$language = $metainfo['language'];
Options::setLanguage($language);
$lastarticlestmp[] = $pageURI->getURI();
$metainfo['oaiidentifier'] = $metainfo['oaiId'];
$metainfo['oaiId'] = Util::getOaiIDfromIdentifier($language, $metainfo['oaiidentifier']);
Logger::info("oaiId " . $metainfo['oaiId']);
//***MAINTAINANCE***
$count++;
//log statistics
if ($count % Options::getOption('printStatInterval') == 0) {
printAll($lastarticles, $language);
}
//50 last articles to statisticdir
if ($count < 50) {
$lastarticles = $lastarticlestmp;
}
if ($count % 50 == 0) {
* the LiveWikipedia. The file outputs the generated triples
* directly. This is the best way for developers to verify that
* their extractors are working. Once the extractor is working on
* an article, developers should use extract_dataset to produce a
* full data set and see whether it works in a full extraction.
*
* See http://wiki.dbpedia.org/Documentation for an overview of
* the DBpedia extraction framework.
*
* @author Jens Lehmann
*/
include 'dbpedia.php';
// configure settings
// change the Extractor class to your extractor
//this should be done in config/dbpedia.ini
Options::setLanguage('ko');
$language = Options::getOption('language');
//$extractor = new ActiveAbstractExtractor();
$extractor = new KoInfoboxExtractor();
/*
$extractor = new InfoboxExtractor();
*/
//$extractor = new SkosCategoriesExtractor();
//these are articles for testing
//$article[] = 'London';
//$article[] = 'Category:Pasta';
$t = '이탈리아';
$t = '서울특별시';
/*
$t = 'Berlin';
*/