function explorerDir($path, $tab_mot_vide)
{
$folder = opendir($path);
while ($entree = readdir($folder)) {
//On ignore les entrées
if ($entree != "." && $entree != "..") {
// On vérifie si il s'agit d'un répertoire
if (is_dir($path . "/" . $entree)) {
$sav_path = $path;
// Construction du path jusqu'au nouveau répertoire
$path .= "/" . $entree;
//echo "DOSSIER = ", $path, "<BR>";
// On parcours le nouveau répertoire
explorerDir($path, $tab_mot_vide);
$path = $sav_path;
} else {
//C'est un fichier html ou pas
$path_source = $path . "/" . $entree;
if (stripos($path_source, '.htm')) {
//echo 'On appelle le module indexation <br>';
//echo $path_source, '<br>';
indexer($path_source, $tab_mot_vide);
}
//Si c'est un .html
//On appelle la fonction d'indexation
//Dans le module_indexation.php
//Par un include
}
}
}
closedir($folder);
}
function summarize($filename, $compress)
{
// proses indexing
$inv_index = indexer();
// load file dan daftar stopwords
$load_file = file_get_contents("./corpus/" . $filename);
$sentence = preg_split("/[.]+/", $load_file);
$sentence = array_slice($sentence, 0, sizeof($sentence) - 1);
// buang array terakhir (kosong)
$stopwords = file_get_contents("./stopwords.txt");
$stopwords = preg_split("/[\\s]+/", $stopwords);
// jumlah kalimat yang diringkas
$compression_rate = $compress / 100;
$max_sentence = floor(sizeof($sentence) * $compression_rate);
// inisialisasi
$sentence_weight = array();
// menghitung bobot tf.idf tiap kalimat
foreach ($sentence as $key => $value) {
// tokenisasi dengan membuang stopwords
$word = preg_split("/[\\d\\W\\s]+/", strtolower($value));
$word = array_diff($word, $stopwords);
$word = array_values($word);
// perbaiki indeks
// inisialisasi bobot dan hitung frekuensi token
$tf_idf = 0;
$freq_word = array_count_values($word);
// hitung bobot tf.idf
foreach ($freq_word as $token => $tf) {
$tf_idf += $tf * $inv_index[$token]['idf'];
}
// simpan nilai bobot kalimat
array_push($sentence_weight, $tf_idf);
}
// sorting bobot tertinggi -> potong array -> sorting urutan kalimat
arsort($sentence_weight);
$sorted = array_slice($sentence_weight, 0, $max_sentence, true);
ksort($sorted);
// gabungkan ringkasan
$summary = "";
foreach ($sorted as $key => $value) {
$summary = $summary . $sentence[$key] . ". ";
}
// return teks asli dan hasil ringkasan
$output = array();
$output['original'] = $load_file;
$output['summary'] = $summary;
return $output;
}
// the index array structure should be (but not comfirmed yet) as below:
// $index = array(
// 'slug-after-hashbang'=>array('timestamp'=>23234234, 'title'=>'Title from that mada', 'excerpt'=>'first para without img'),
// 'slug-after-hashbang'=>array('timestamp'=>23234234, 'title'=>'Title from that mada', 'excerpt'=>'first para without img'),
// 'slug-after-hashbang'=>array('timestamp'=>23234234, 'title'=>'Title from that mada', 'excerpt'=>'first para without img'),
// 'slug-after-hashbang'=>array('timestamp'=>23234234, 'title'=>'Title from that mada', 'excerpt'=>'first para without img'),
// );
// in json mind you. :)
## Configurations:
const MARKDOWN_DIR = 'posts';
error_reporting(E_ALL);
## MAIN
$mds = get_mds();
foreach ($mds as $filename) {
if ($filename != '.DS_Store') {
$ind = indexer($filename);
$index[$ind['slug']] = $ind;
}
}
dumper($index);
$data = json_encode($index);
file_put_contents('assets/index.json', $data);
echo 'done!';
#################################
########### FUNCTIONS ###########
function dumper($multi)
{
echo '<pre>';
var_dump($multi);
echo '</pre>';
}
