function scrape()
{
$riga = 2;
$i = 0;
do {
//scorre le righe
$filter = "//table[2]/tr[{$riga}]/th[1]";
parent::XPathFilter($filter);
$tmp = parent::toString('txt');
if ($tmp == "") {
continue;
}
$this->siti[$i] = $tmp;
$filter = "//table[2]/tr[{$riga}]/th[2]";
parent::XPathFilter($filter);
$tmp = parent::toString('txt');
$this->nomi[$i] = $tmp;
$riga++;
$i++;
} while ($tmp != "");
$out = "";
foreach ($this->nomi as $k => $v) {
$a = trim($this->nomi[$k]);
$b = trim($this->siti[$k]);
$out .= "{\"nome\" : \"{$a}\",\"sito\" : \"{$b}\"},";
}
$out = substr($out, 0, -1);
return "[" . $out . "]";
}
function unibo()
{
$this->location = "div1_div3_div2";
//titolo
parent::XPathFilter("//div[@id=\"articleTitle\"]/h3");
$this->title = parent::toString("txt");
$this->title = trim($this->title);
$this->locations["title"] = "_div3_div2_h31";
$this->start_array["title"] = 0;
$this->end_array["title"] = mb_strlen($this->title, "UTF-8");
//anno pubblicazione
parent::XPathFilter("//div[@id=\"breadcrumb\"]//a[@target=\"_parent\"][2]");
$this->year = parent::toString("txt");
$arr = explode(" ", $this->year);
$arr2 = explode("(", $this->year);
$this->year = $arr[count($arr) - 1];
$this->year = substr($this->year, 1, 4);
$this->locations["year"] = "_div2_a2";
$start = mb_strlen($arr2[0], "UTF-8") + 1;
$this->start_array["year"] = $start;
$this->end_array["year"] = $start + mb_strlen($this->year, "UTF-8");
//autori
parent::XPathFilter("//div[@id=\"authorString\"]/em");
$autore = parent::toString("txt");
$this->authors = explode(", ", $autore);
if (strcmp($this->authors[0], "") == 0) {
//dato che l'explode di una stringa vuota da come risultato un array di un elemento con chiave 0 e valore stringa vuota è necessario questo controllo
$this->authors = null;
} else {
$this->locations["authors"] = "_div3_div3_em1";
$start = 0;
$end = 0;
foreach ($this->authors as $key => $val) {
$this->authors[$key] = trim($val);
$val = $this->authors[$key];
if ($key == 0) {
$start = 0;
$end = mb_strlen($val, "UTF-8");
$this->start_array[$key] = $start;
$this->end_array[$key] = $end;
} else {
$start = $end + 2;
$end = $start + mb_strlen($val, "UTF-8");
$this->start_array[$key] = $start;
$this->end_array[$key] = $end;
}
}
//inutili per unibo, ma servono per dare uniformità con dlib
$this->locations["firstauthors"] = "";
$this->locations["continuation_of_authors"] = "";
}
//doi
parent::XPathFilter("//a[@id=\"pub-id::doi\"]");
$this->doi = parent::toString("txt");
$this->doi = trim($this->doi);
$this->locations["doi"] = "_div3_a1";
$this->start_array["doi"] = 0;
$this->end_array["doi"] = mb_strlen($this->doi, "UTF-8");
//abstract
parent::XPathFilter("//div[@id=\"articleAbstract\"]//div");
$this->abstract = parent::toString("txt");
$this->abstract = trim($this->abstract);
$tmp = parent::toString();
//prendo il nodo con ancora i tag html per controllare se il contenuto dell' abstract direttamente nel div o in un p dentro al div
$tmp = substr($tmp, 5, 3);
$this->locations["abstract"] = "_div3_div4_div1";
if (strpos($tmp, "<p") !== FALSE) {
$this->locations["abstract"] .= "_p1";
}
$this->start_array["abstract"] = 0;
$this->end_array["abstract"] = mb_strlen($this->abstract, "UTF-8");
//su unibo le intro sono al più dentro i documenti pdf, quindi non è richiesto che lo scraper le sappia prendere
//citazioni
$set = null;
parent::XPathFilter("//body//div[@id=\"articleCitations\"]/div[1]//p");
$set = $this->nodes->each(function ($node) {
return $node->text();
});
$citazione = null;
foreach ($set as $nann => $v) {
$title = null;
$year = null;
$doi = null;
$url = null;
$authors = array();
$tmp = 0;
//target
$pos = $nann + 1;
$this->locations["citos"][$nann] = "_div3_div7_div1_p{$pos}";
$this->start_array["citos"][$nann] = 0;
$this->end_array["citos"][$nann] = $this->start_array["citos"][$nann] + mb_strlen(trim($v), "UTF-8");
//autori
$authors = getAuthorsUnibo($v);
//anno della citazione
$year = getYear($v, $this->year);
//titolo
$title = getTitleUnibo($v, $year, $authors);
//doi e url della citazione
$coppia = getDoiAndURL($v);
if (isset($coppia[0])) {
$url = $coppia[0];
}
if (isset($coppia[1])) {
$doi = $coppia[1];
}
$this->citations[$nann] = new citationcontainer($v, $title, $year, $authors, $doi, $url);
/*echo "<br><br> Annotazione $nann:<br>";
echo "Titolo trovato: $title<br>";
echo "Anno trovato: $year<br>";
echo "Url trovato: $url<br>";
echo "Doi trovato: $doi<br>";
echo "Lista autori:<br>";
foreach ($authors as $k=>$val){
echo "Autore $k: $val<br>";
}
echo "Target:<br>";
echo "location: ".$this->location.$this->locations["citos"][$nann]."<br>";
echo "start: ".$this->start_array["citos"][$nann]."<br>";
echo "end: ".$this->end_array["citos"][$nann]."<br>";
echo "<br>";
*/
}
}
<?php
//scritto da Davide Quadrelli
header("Content-Type:html;charset=UTF-8");
require_once 'include/Scraper.php';
$toret = array();
if (isset($_GET['url'])) {
$scraper = new Scraper();
$scraper->loadPage($_GET["url"], true);
if (strpos($_GET['url'], "dlib")) {
//pagina di d-lib
$scraper->XPathFilter("html/body[1]/form/table[3]/tr[1]/td[1]/table[5]/tr[1]/td[1]/table[1]/tr[1]/td[2]/node()");
$toret[0] = "form1_table3_tr1_td1_table5_tr1_td1_table1_tr1_td2_";
} else {
if (strpos($_GET['url'], "unibo")) {
//articolo di almajournal
$toret[0] = "div1_div3_div2_";
/*$scraper->XPathFilter("//div[@id=\"articleTitle\"] | //div[@id=\"authorString\"] | //div[@id=\"articleAbstract\"] | //div[@id=\"articleSubject\"]
| //div[@id=\"articleFullText\"] | //div[@id=\"authorString\"] | //div[@id=\"articleCitations\"] | //a[@id=\"pub-id::doi\"]");*/
$scraper->XPathFilter("//div[@id=\"main\"]/node()");
if ($scraper->toString() == "") {
//è un sito unibo ma non un articolo
$toret[0] = "body1_";
$scraper->XPathFilter("//body");
}
} else {
$toret[0] = "body1_";
$scraper->XPathFilter("//body");
}
}
$toret[1] = $scraper->toString();
