191 lines
7.4 KiB
Java
191 lines
7.4 KiB
Java
package eu.dnetlib.data.collector.plugins.httpfilename;
|
|
|
|
import java.util.*;
|
|
import java.util.concurrent.ArrayBlockingQueue;
|
|
import java.util.concurrent.TimeUnit;
|
|
|
|
import eu.dnetlib.data.collector.rmi.CollectorServiceException;
|
|
import org.apache.commons.logging.Log;
|
|
import org.apache.commons.logging.LogFactory;
|
|
import org.json.JSONObject;
|
|
import org.json.XML;
|
|
import org.jsoup.Jsoup;
|
|
import org.jsoup.nodes.Document;
|
|
import org.jsoup.nodes.Element;
|
|
import org.jsoup.select.Elements;
|
|
|
|
/**
|
|
* Created by miriam on 04/05/2018.
|
|
*/
|
|
public class HTTPWithFileNameCollectorIterable implements Iterable<String> {
|
|
|
|
private static final Log log = LogFactory.getLog(HTTPWithFileNameCollectorIterable.class);
|
|
|
|
private static final String JUNK = "<resource><url>%s</url><DOI>JUNK</DOI></resource>";
|
|
public static final String APP_JSON = "application/json";
|
|
public static final String APP_XML = "application/xml";
|
|
public static final String TEXT_HTML = "text/html";
|
|
private final ArrayBlockingQueue<String> queue = new ArrayBlockingQueue<String>(100);
|
|
|
|
|
|
|
|
|
|
private String filterParam;
|
|
|
|
int total = 0;
|
|
int filtered = 0;
|
|
|
|
public HTTPWithFileNameCollectorIterable(String startUrl, String filter){
|
|
|
|
this.filterParam = filter;
|
|
Thread ft = new Thread(new FillMetaQueue(startUrl) );
|
|
ft.start();
|
|
}
|
|
|
|
|
|
@Override
|
|
public Iterator<String> iterator() {
|
|
return new HttpWithFileNameCollectorIterator(queue);
|
|
}
|
|
|
|
private class FillMetaQueue implements Runnable {
|
|
final Connector c = new Connector();
|
|
|
|
private final List<String> metas = Collections.synchronizedList(new ArrayList<String>());
|
|
private final List<String> urls = Collections.synchronizedList(new ArrayList<>());
|
|
|
|
public FillMetaQueue(String startUrl){
|
|
if(!startUrl.isEmpty()){
|
|
urls.add(startUrl);
|
|
}
|
|
}
|
|
|
|
|
|
public void fillQueue() {
|
|
String url;
|
|
|
|
while((metas.size()>0 || urls.size() > 0 )) {
|
|
log.debug("metas.size() = " + metas.size() + " urls.size() = " + urls.size() + " queue.size() = " +queue.size());
|
|
if (metas.size() > 0) {
|
|
url = metas.remove(0);
|
|
try {
|
|
c.get(url);
|
|
} catch (CollectorServiceException e) {
|
|
log.info("Impossible to collect url: " + url + " error: " + e.getMessage());
|
|
}
|
|
if(c.isStatusOk()){
|
|
try {
|
|
String ret = c.getResponse();
|
|
if (ret != null && ret.length()>0) {
|
|
if (!containsFilter(ret))
|
|
queue.put(addFilePath(ret, url, url.endsWith(".json")));
|
|
//queue.offer(addFilePath(ret, url, url.endsWith(".json")), HttpWithFileNameCollectorIterator.waitTime, TimeUnit.SECONDS);
|
|
else
|
|
filtered++;
|
|
total++;
|
|
}
|
|
} catch (InterruptedException e) {
|
|
log.info("not inserted in queue element associate to url " + url + " error: " + e.getMessage() );
|
|
|
|
}
|
|
}
|
|
} else {
|
|
url = urls.remove(0);
|
|
try {
|
|
c.get(url);
|
|
} catch (CollectorServiceException e) {
|
|
log.info("Impossible to collect url: " + url + " error: " + e.getMessage());
|
|
}
|
|
if(c.isStatusOk()) {
|
|
if (c.responseTypeContains(TEXT_HTML)){
|
|
recurFolder(c.getResponse(), url);
|
|
} else if(c.responseTypeContains(APP_JSON) || c.responseTypeContains(APP_XML)){
|
|
try {
|
|
final String element = addFilePath(c.getResponse(), url, c.responseTypeContains(APP_JSON));
|
|
//queue.offer(element, HttpWithFileNameCollectorIterator.waitTime, TimeUnit.SECONDS);
|
|
queue.put(element);
|
|
} catch (InterruptedException e) {
|
|
log.info("not inserted in queue element associate to url " + url + " error: " + e.getMessage() );
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
}
|
|
try {
|
|
//queue.offer(HttpWithFileNameCollectorIterator.TERMINATOR, HttpWithFileNameCollectorIterator.waitTime, TimeUnit.SECONDS);
|
|
queue.put(HttpWithFileNameCollectorIterator.TERMINATOR);
|
|
} catch (InterruptedException e) {
|
|
throw new IllegalStateException(String.format("could not add element to queue for more than %s%s", HttpWithFileNameCollectorIterator.waitTime, TimeUnit.SECONDS), e);
|
|
}
|
|
|
|
}
|
|
|
|
private boolean containsFilter(String meta){
|
|
if (filterParam == null || filterParam.isEmpty())
|
|
return false;
|
|
String[] filter = filterParam.split(";");
|
|
for(String item:filter){
|
|
if (meta.contains(item))
|
|
return true;
|
|
}
|
|
return false;
|
|
}
|
|
|
|
private String addFilePath(String meta, String url, boolean isJson){
|
|
String path = url.replace("metadata", "pdf");
|
|
|
|
try {
|
|
if(isJson)
|
|
meta = meta.substring(0, meta.length() - 1) + ",'downloadFileUrl':'" + path.substring(0, path.indexOf(".json")) + ".pdf'}";
|
|
else {
|
|
|
|
if (meta.contains("<!DOCTYPE")) {
|
|
meta = meta.substring(meta.indexOf("<!DOCTYPE"));
|
|
meta = meta.substring(meta.indexOf(">") + 1);
|
|
}
|
|
int index = meta.lastIndexOf("</");
|
|
meta = meta.substring(0, index) + "<downloadFileUrl>" + path.substring(0, path.indexOf(".xml")) + ".pdf</downloadFileUrl>" + meta.substring(index);
|
|
}
|
|
} catch(Exception ex) {
|
|
log.info("not file with extension .json or .xml");
|
|
}
|
|
|
|
|
|
if(isJson) {
|
|
try {
|
|
return XML.toString(new JSONObject("{'resource':" + meta + "}"));
|
|
} catch(Exception e) {
|
|
log.fatal("Impossible to transform json object to xml \n" + meta + "\n " + e.getMessage() + "\n" + url);
|
|
// throw new RuntimeException();
|
|
final String junk = String.format(JUNK, url);
|
|
log.warn("returning " + junk);
|
|
return junk;
|
|
}
|
|
}
|
|
return meta;
|
|
}
|
|
|
|
private void recurFolder(String text, String url){
|
|
Document doc = Jsoup.parse(text);
|
|
Elements links = doc.select("a");
|
|
for(Element e:links){
|
|
if (!e.text().equals("../")){
|
|
String file = e.attr("href");
|
|
if(file.endsWith(".json") || file.endsWith(".xml"))
|
|
metas.add(url+file);
|
|
else
|
|
urls.add(url+file);
|
|
}
|
|
}
|
|
}
|
|
|
|
|
|
@Override
|
|
public void run() {
|
|
fillQueue();
|
|
}
|
|
}
|
|
|
|
}
|