yacy_search_server/htroot/rct_p.java

129 lines
5.1 KiB
Java
Raw Normal View History

// rct_p.java
// -----------------------
// (C) 2007 by Michael Peter Christen; mc@yacy.net, Frankfurt a. M., Germany
// first published 28.11.2007 on http://yacy.net
//
// This is a part of YaCy, a peer-to-peer based web search engine
//
// $LastChangedDate$
// $LastChangedRevision$
// $LastChangedBy$
//
// LICENSE
//
// This program is free software; you can redistribute it and/or modify
// it under the terms of the GNU General Public License as published by
// the Free Software Foundation; either version 2 of the License, or
// (at your option) any later version.
//
// This program is distributed in the hope that it will be useful,
// but WITHOUT ANY WARRANTY; without even the implied warranty of
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
// GNU General Public License for more details.
//
// You should have received a copy of the GNU General Public License
// along with this program; if not, write to the Free Software
// Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
import java.net.MalformedURLException;
import java.util.Date;
import java.util.Iterator;
import net.yacy.cora.document.Hit;
import net.yacy.cora.document.RSSFeed;
import net.yacy.cora.protocol.RequestHeader;
import net.yacy.kelondro.data.meta.DigestURI;
import net.yacy.peers.Protocol;
import net.yacy.peers.Seed;
import net.yacy.peers.dht.PeerSelection;
import net.yacy.search.Switchboard;
import de.anomic.crawler.retrieval.Request;
import de.anomic.server.serverObjects;
import de.anomic.server.serverSwitch;
public class rct_p {
public static serverObjects respond(@SuppressWarnings("unused") final RequestHeader header, final serverObjects post, final serverSwitch env) {
// return variable that accumulates replacements
final Switchboard sb = (Switchboard) env;
final serverObjects prop = new serverObjects();
if (post != null) {
if (post.containsKey("retrieve")) {
final String peerhash = post.get("peer", null);
final Seed seed = (peerhash == null) ? null : sb.peers.getConnected(peerhash);
final RSSFeed feed = (seed == null) ? null : Protocol.queryRemoteCrawlURLs(sb.peers, seed, 20, 60000);
if (feed != null) {
for (final Hit item: feed) {
//System.out.println("URL=" + item.getLink() + ", desc=" + item.getDescription() + ", pubDate=" + item.getPubDate());
// put url on remote crawl stack
DigestURI url;
try {
url = new DigestURI(item.getLink());
} catch (final MalformedURLException e) {
url = null;
}
Date loaddate;
loaddate = item.getPubDate();
final DigestURI referrer = null; // referrer needed!
final String urlRejectReason = sb.crawlStacker.urlInAcceptedDomain(url);
if (urlRejectReason == null) {
// stack url
if (sb.getLog().isFinest()) sb.getLog().logFinest("crawlOrder: stack: url='" + url + "'");
sb.crawlStacker.enqueueEntry(new Request(
peerhash.getBytes(),
url,
(referrer == null) ? null : referrer.hash(),
"REMOTE-CRAWLING",
loaddate,
sb.crawler.defaultRemoteProfile.handle(),
0,
0,
0,
item.getSize()
));
} else {
env.getLog().logWarning("crawlOrder: Rejected URL '" + urlToString(url) + "': " + urlRejectReason);
}
}
}
}
}
major step forward to network switching (target is easy switch to intranet or other networks .. and back) This change is inspired by the need to see a network connected to the index it creates in a indexing team. It is not possible to divide the network and the index. Therefore all control files for the network was moved to the network within the INDEX/<network-name> subfolder. The remaining YACYDB is superfluous and can be deleted. The yacyDB and yacyNews data structures are now part of plasmaWordIndex. Therefore all methods, using static access to yacySeedDB had to be rewritten. A special problem had been all the port forwarding methods which had been tightly mixed with seed construction. It was not possible to move the port forwarding functions to the place, meaning and usage of plasmaWordIndex. Therefore the port forwarding had been deleted (I guess nobody used it and it can be simulated by methods outside of YaCy). The mySeed.txt is automatically moved to the current network position. A new effect causes that every network will create a different local seed file, which is ok, since the seed identifies the peer only against the network (it is the purpose of the seed hash to give a peer a location within the DHT). No other functional change has been made. The next steps to enable network switcing are: - shift of crawler tables from PLASMADB into the network (crawls are also network-specific) - possibly shift of plasmaWordIndex code into yacy package (index management is network-specific) - servlet to switch networks git-svn-id: https://svn.berlios.de/svnroot/repos/yacy/trunk@4765 6c8d7289-2bf4-0310-a012-ef5d649a1542
2008-05-06 01:13:47 +02:00
listHosts(sb, prop);
// return rewrite properties
return prop;
}
/**
* @param url
* @return
*/
private static String urlToString(final DigestURI url) {
return (url == null ? "null" : url.toNormalform(true, false));
}
private static void listHosts(final Switchboard sb, final serverObjects prop) {
// list known hosts
Seed seed;
int hc = 0;
if (sb.peers != null && sb.peers.sizeConnected() > 0) {
final Iterator<Seed> e = PeerSelection.getProvidesRemoteCrawlURLs(sb.peers);
while (e.hasNext()) {
seed = e.next();
if (seed != null) {
prop.put("hosts_" + hc + "_hosthash", seed.hash);
prop.putHTML("hosts_" + hc + "_hostname", seed.hash + " " + seed.get(Seed.NAME, "nameless") + " (" + seed.getLong(Seed.RCOUNT, 0) + ")");
hc++;
}
}
prop.put("hosts", hc);
} else {
prop.put("hosts", "0");
}
}
}