mirror of
https://github.com/yacy/yacy_search_server.git
synced 2024-09-21 00:00:13 +02:00
7535fd7447
- introduced blocking queues in CrawlStacker to make it ready for concurrency - added a second busy thread for the CrawlStacker The CrawlStacker is multithreaded. It shall be transformed into a BlockingThread in another step. The concurrency of the stacker will hopefully solve some problems with cases where DNS blocks. git-svn-id: https://svn.berlios.de/svnroot/repos/yacy/trunk@5395 6c8d7289-2bf4-0310-a012-ef5d649a1542
370 lines
16 KiB
Java
Executable File
370 lines
16 KiB
Java
Executable File
// CrawlEntry.java
|
|
// (C) 2007 by Michael Peter Christen; mc@yacy.net, Frankfurt a. M., Germany
|
|
// first published 14.03.2007 on http://yacy.net
|
|
//
|
|
// This is a part of YaCy, a peer-to-peer based web search engine
|
|
//
|
|
// $LastChangedDate: 2006-04-02 22:40:07 +0200 (So, 02 Apr 2006) $
|
|
// $LastChangedRevision: 1986 $
|
|
// $LastChangedBy: orbiter $
|
|
//
|
|
// 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
|
|
|
|
package de.anomic.crawler;
|
|
|
|
import java.io.IOException;
|
|
import java.io.UnsupportedEncodingException;
|
|
import java.util.Date;
|
|
import java.util.concurrent.ConcurrentHashMap;
|
|
|
|
import de.anomic.kelondro.kelondroBase64Order;
|
|
import de.anomic.kelondro.kelondroBitfield;
|
|
import de.anomic.kelondro.kelondroNaturalOrder;
|
|
import de.anomic.kelondro.kelondroRow;
|
|
import de.anomic.plasma.plasmaSwitchboard;
|
|
import de.anomic.server.serverProcessorJob;
|
|
import de.anomic.yacy.yacySeedDB;
|
|
import de.anomic.yacy.yacyURL;
|
|
|
|
public class CrawlEntry extends serverProcessorJob {
|
|
|
|
// row definition for balancer-related NURL-entries
|
|
public final static kelondroRow rowdef = new kelondroRow(
|
|
"String urlhash-" + yacySeedDB.commonHashLength + ", " + // the url's hash
|
|
"String initiator-" + yacySeedDB.commonHashLength + ", " + // the crawling initiator
|
|
"String urlstring-256, " + // the url as string
|
|
"String refhash-" + yacySeedDB.commonHashLength + ", " + // the url's referrer hash
|
|
"String urlname-80, " + // the name of the url, from anchor tag <a>name</a>
|
|
"Cardinal appdate-8 {b256}, " + // the time when the url was first time appeared
|
|
"String profile-" + yacySeedDB.commonHashLength + ", " + // the name of the prefetch profile handle
|
|
"Cardinal depth-2 {b256}, " + // the prefetch depth so far, starts at 0
|
|
"Cardinal parentbr-3 {b256}, " + // number of anchors of the parent
|
|
"Cardinal forkfactor-4 {b256}, " + // sum of anchors of all ancestors
|
|
"byte[] flags-4, " + // flags
|
|
"String handle-4, " + // extra handle
|
|
"Cardinal loaddate-8 {b256}," + // time when the file was loaded
|
|
"Cardinal serverdate-8 {b256}," + // time when that the server returned as document date
|
|
"Cardinal modifiedSince-8 {b256}", // time that was given to server as ifModifiedSince
|
|
kelondroBase64Order.enhancedCoder,
|
|
0
|
|
);
|
|
|
|
// a shared domainAccess map for all balancers. the key is a domain-hash (6 bytes)
|
|
public static final ConcurrentHashMap<String, domaccess> domainAccess = new ConcurrentHashMap<String, domaccess>();
|
|
|
|
private String initiator; // the initiator hash, is NULL or "" if it is the own proxy;
|
|
// if this is generated by a crawl, the own peer hash in entered
|
|
private String refhash; // the url's referrer hash
|
|
private yacyURL url; // the url as string
|
|
private String name; // the name of the url, from anchor tag <a>name</a>
|
|
private long appdate; // the time when the url was first time appeared
|
|
private long loaddate; // the time when the url was loaded
|
|
private long serverdate; // the document date from the target server
|
|
private long imsdate; // the time of a ifModifiedSince request
|
|
private String profileHandle; // the name of the prefetch profile
|
|
private int depth; // the prefetch depth so far, starts at 0
|
|
private int anchors; // number of anchors of the parent
|
|
private int forkfactor; // sum of anchors of all ancestors
|
|
private kelondroBitfield flags;
|
|
private int handle;
|
|
private String statusMessage;
|
|
private int initialHash; // to provide a object hash that does not change even if the url changes because of redirection
|
|
|
|
public static class domaccess {
|
|
public long time;
|
|
public long robotsMinDelay;
|
|
public int count;
|
|
public String host;
|
|
public domaccess(String host) {
|
|
this.host = host;
|
|
this.time = System.currentTimeMillis();
|
|
this.robotsMinDelay = 0;
|
|
this.count = 0;
|
|
}
|
|
public void update() {
|
|
this.time = System.currentTimeMillis();
|
|
this.count++;
|
|
}
|
|
public long flux(long range) {
|
|
return count >= 1000 ? range * Math.min(5000, count) / 1000 : range / (1000 - count);
|
|
}
|
|
}
|
|
|
|
|
|
/**
|
|
* @param initiator the hash of the initiator peer
|
|
* @param url the {@link URL} to crawl
|
|
* @param referrer the hash of the referrer URL
|
|
* @param name the name of the document to crawl
|
|
* @param appdate the time when the url was first time appeared
|
|
* @param profileHandle the name of the prefetch profile. This must not be null!
|
|
* @param depth the crawling depth of the entry
|
|
* @param anchors number of anchors of the parent
|
|
* @param forkfactor sum of anchors of all ancestors
|
|
*/
|
|
public CrawlEntry(
|
|
final String initiator,
|
|
final yacyURL url,
|
|
final String referrerhash,
|
|
final String name,
|
|
final Date appdate,
|
|
final Date loaddate,
|
|
final String profileHandle,
|
|
final int depth,
|
|
final int anchors,
|
|
final int forkfactor
|
|
) {
|
|
// create new entry and store it into database
|
|
assert url != null;
|
|
assert initiator != null;
|
|
assert profileHandle.length() == yacySeedDB.commonHashLength : profileHandle + " != " + yacySeedDB.commonHashLength;
|
|
this.initiator = initiator;
|
|
this.url = url;
|
|
this.refhash = (referrerhash == null) ? "" : referrerhash;
|
|
this.name = (name == null) ? "" : name;
|
|
this.appdate = (appdate == null) ? 0 : appdate.getTime();
|
|
this.loaddate = (loaddate == null) ? 0 : loaddate.getTime();
|
|
this.profileHandle = profileHandle; // must not be null
|
|
this.depth = depth;
|
|
this.anchors = anchors;
|
|
this.forkfactor = forkfactor;
|
|
this.flags = new kelondroBitfield(rowdef.width(10));
|
|
this.handle = 0;
|
|
this.serverdate = 0;
|
|
this.imsdate = 0;
|
|
this.statusMessage = "loaded(args)";
|
|
this.initialHash = url.hashCode();
|
|
this.status = serverProcessorJob.STATUS_INITIATED;
|
|
}
|
|
|
|
public CrawlEntry(final kelondroRow.Entry entry) throws IOException {
|
|
assert (entry != null);
|
|
insertEntry(entry);
|
|
}
|
|
|
|
private void insertEntry(final kelondroRow.Entry entry) throws IOException {
|
|
final String urlstring = entry.getColString(2, null);
|
|
if (urlstring == null) throw new IOException ("url string is null");
|
|
this.initiator = entry.getColString(1, null);
|
|
this.url = new yacyURL(urlstring, entry.getColString(0, null));
|
|
this.refhash = (entry.empty(3)) ? "" : entry.getColString(3, null);
|
|
this.name = (entry.empty(4)) ? "" : entry.getColString(4, "UTF-8").trim();
|
|
this.appdate = entry.getColLong(5);
|
|
this.profileHandle = (entry.empty(6)) ? null : entry.getColString(6, null).trim();
|
|
this.depth = (int) entry.getColLong(7);
|
|
this.anchors = (int) entry.getColLong(8);
|
|
this.forkfactor = (int) entry.getColLong(9);
|
|
this.flags = new kelondroBitfield(entry.getColBytes(10));
|
|
this.handle = Integer.parseInt(entry.getColString(11, null), 16);
|
|
this.loaddate = entry.getColLong(12);
|
|
this.serverdate = entry.getColLong(13);
|
|
this.imsdate = entry.getColLong(14);
|
|
this.statusMessage = "loaded(kelondroRow.Entry)";
|
|
this.initialHash = url.hashCode();
|
|
return;
|
|
}
|
|
|
|
public int hashCode() {
|
|
// overloads Object.hashCode()
|
|
return this.initialHash;
|
|
}
|
|
|
|
public void setStatus(final String s, int code) {
|
|
this.statusMessage = s;
|
|
this.status = code;
|
|
}
|
|
|
|
public String getStatus() {
|
|
return this.statusMessage;
|
|
}
|
|
|
|
private static String normalizeHandle(final int h) {
|
|
String d = Integer.toHexString(h);
|
|
while (d.length() < rowdef.width(11)) d = "0" + d;
|
|
return d;
|
|
}
|
|
|
|
public kelondroRow.Entry toRow() {
|
|
final byte[] appdatestr = kelondroNaturalOrder.encodeLong(appdate, rowdef.width(5));
|
|
final byte[] loaddatestr = kelondroNaturalOrder.encodeLong(loaddate, rowdef.width(12));
|
|
final byte[] serverdatestr = kelondroNaturalOrder.encodeLong(serverdate, rowdef.width(13));
|
|
final byte[] imsdatestr = kelondroNaturalOrder.encodeLong(imsdate, rowdef.width(14));
|
|
// store the hash in the hash cache
|
|
byte[] namebytes;
|
|
try {
|
|
namebytes = this.name.getBytes("UTF-8");
|
|
} catch (final UnsupportedEncodingException e) {
|
|
namebytes = this.name.getBytes();
|
|
}
|
|
final byte[][] entry = new byte[][] {
|
|
this.url.hash().getBytes(),
|
|
(initiator == null) ? "".getBytes() : this.initiator.getBytes(),
|
|
this.url.toString().getBytes(),
|
|
this.refhash.getBytes(),
|
|
namebytes,
|
|
appdatestr,
|
|
(this.profileHandle == null) ? null : this.profileHandle.getBytes(),
|
|
kelondroNaturalOrder.encodeLong(this.depth, rowdef.width(7)),
|
|
kelondroNaturalOrder.encodeLong(this.anchors, rowdef.width(8)),
|
|
kelondroNaturalOrder.encodeLong(this.forkfactor, rowdef.width(9)),
|
|
this.flags.bytes(),
|
|
normalizeHandle(this.handle).getBytes(),
|
|
loaddatestr,
|
|
serverdatestr,
|
|
imsdatestr};
|
|
return rowdef.newEntry(entry);
|
|
}
|
|
|
|
public yacyURL url() {
|
|
// the url
|
|
return url;
|
|
}
|
|
|
|
public void redirectURL(final yacyURL redirectedURL) {
|
|
// replace old URL by new one. This should only be used in case of url redirection
|
|
this.url = redirectedURL;
|
|
}
|
|
|
|
public String referrerhash() {
|
|
// the urlhash of a referer url
|
|
return this.refhash;
|
|
}
|
|
|
|
public String initiator() {
|
|
// returns the hash of the initiating peer
|
|
if (initiator == null) return "";
|
|
if (initiator.length() == 0) return "";
|
|
return initiator;
|
|
}
|
|
|
|
public boolean proxy() {
|
|
// true when the url was retrieved using the proxy
|
|
return (initiator() == null);
|
|
}
|
|
|
|
public Date appdate() {
|
|
// the date when the url appeared first
|
|
return new Date(this.appdate);
|
|
}
|
|
|
|
public Date loaddate() {
|
|
// the date when the url was loaded
|
|
return new Date(this.loaddate);
|
|
}
|
|
|
|
public Date serverdate() {
|
|
// the date that the server returned as document date
|
|
return new Date(this.serverdate);
|
|
}
|
|
|
|
public Date imsdate() {
|
|
// the date that the client (browser) send as ifModifiedSince in proxy mode
|
|
return new Date(this.imsdate);
|
|
}
|
|
|
|
public String name() {
|
|
// return the anchor name (text inside <a> tag)
|
|
return this.name;
|
|
}
|
|
|
|
public int depth() {
|
|
// crawl depth where the url appeared
|
|
return this.depth;
|
|
}
|
|
|
|
public String profileHandle() {
|
|
// the handle of the crawl profile
|
|
assert profileHandle.length() == yacySeedDB.commonHashLength : profileHandle + " != " + yacySeedDB.commonHashLength;
|
|
return this.profileHandle;
|
|
}
|
|
|
|
/**
|
|
* check a domain flag so it can be calculated when a domain was accessed the last time
|
|
*/
|
|
public void updateAccess() {
|
|
String domhash = url.hash().substring(6);
|
|
domaccess lastAccess = domainAccess.get(domhash);
|
|
if (lastAccess == null) {
|
|
lastAccess = new domaccess(url.getHost());
|
|
domainAccess.put(domhash, lastAccess);
|
|
} else {
|
|
lastAccess.update();
|
|
}
|
|
}
|
|
|
|
/**
|
|
* calculates how long should be waited until the domain can be accessed again
|
|
* this follows from given minimum access times, the fact that an url is a CGI url or now, the times that the domain was accessed
|
|
* and a given minimum access time as given in robots.txt
|
|
* @param minimumLocalDelta
|
|
* @param minimumGlobalDelta
|
|
* @return the remaining waiting time in milliseconds
|
|
*/
|
|
public long waitingRemaining(final long minimumLocalDelta, final long minimumGlobalDelta) {
|
|
final long delta = lastAccessDelta(this.url.hash());
|
|
if (delta == Long.MAX_VALUE) return 0;
|
|
final boolean local = this.url.isLocal();
|
|
long deltaBase = (local) ? minimumLocalDelta : minimumGlobalDelta;
|
|
if (this.url.isCGI()) deltaBase = deltaBase * 2; // mostly there is a database access in the background which creates a lot of unwanted IO on target site
|
|
domaccess lastAccess = domainAccess.get(this.url.hash().substring(6));
|
|
lastAccess.robotsMinDelay = (local) ? 0 : plasmaSwitchboard.getSwitchboard().robots.crawlDelayMillis(this.url);
|
|
final long genericDelta = Math.min(
|
|
60000,
|
|
Math.max(
|
|
deltaBase + ((lastAccess == null || local) ? 0 : lastAccess.flux(deltaBase)),
|
|
(local || lastAccess == null) ? 0 : lastAccess.robotsMinDelay)
|
|
); // prevent that that robots file can stop our indexer completely
|
|
return (delta < genericDelta) ? genericDelta - delta : 0;
|
|
}
|
|
|
|
/**
|
|
* guess a minimum waiting time
|
|
* the time is not correct, because if the domain was not checked yet by the robots.txt delay value, it is too low
|
|
* also the 'isCGI' property is missing, because the full text of the domain is unknown here
|
|
* @param urlhash
|
|
* @param minimumLocalDelta
|
|
* @param minimumGlobalDelta
|
|
* @return the remaining waiting time in milliseconds
|
|
*/
|
|
public static long waitingRemainingGuessed(String urlhash, final long minimumLocalDelta, final long minimumGlobalDelta) {
|
|
final long delta = lastAccessDelta(urlhash);
|
|
if (delta == Long.MAX_VALUE) return 0;
|
|
final boolean local = yacyURL.isLocal(urlhash);
|
|
long deltaBase = (local) ? minimumLocalDelta : minimumGlobalDelta;
|
|
domaccess lastAccess = domainAccess.get(urlhash.substring(6));
|
|
final long genericDelta = Math.min(
|
|
60000,
|
|
Math.max(
|
|
deltaBase + ((lastAccess == null || local) ? 0 : lastAccess.flux(deltaBase)),
|
|
(local || lastAccess == null) ? 0 : lastAccess.robotsMinDelay)
|
|
); // prevent that that robots file can stop our indexer completely
|
|
return (delta < genericDelta) ? genericDelta - delta : 0;
|
|
}
|
|
|
|
/**
|
|
* calculates the time since the last access of the domain as referenced by the url hash
|
|
* @param urlhash
|
|
* @return a time in milliseconds since last access of the domain or Long.MAX_VALUE if the domain was not accessed before
|
|
*/
|
|
private static long lastAccessDelta(final String hash) {
|
|
assert hash != null;
|
|
assert hash.length() == 6 || hash.length() == 12;
|
|
final domaccess lastAccess = domainAccess.get((hash.length() > 6) ? hash.substring(6) : hash);
|
|
if (lastAccess == null) return Long.MAX_VALUE; // never accessed
|
|
return System.currentTimeMillis() - lastAccess.time;
|
|
}
|
|
} |