yacy_search_server/source/de/anomic/plasma/plasmaSwitchboardQueue.java
orbiter a31b9097a4 preparations for mass remote crawls:
two main changes must be implemented to enable mass remote crawls:
- shift control of robots.txt to crawl queue (away from stacker). This is necessary since remote
  crawls can contain unchecked urls. Each peer must check the robots to prevent that it is misused
  as crawl agent for unwanted file retrieval
- implement new index files that control double-check of remotely crawled urls

After removal of robots.txt checking from stacker threads, the multi-threading of this process is void.
Multithreading has been removed. Also the thread pools for the crawl threads had been removed, since
creation of these threads is not resource-consuming, for a detailed explanation see svn 4106

git-svn-id: https://svn.berlios.de/svnroot/repos/yacy/trunk@4181 6c8d7289-2bf4-0310-a012-ef5d649a1542
2007-10-29 01:43:20 +00:00

473 lines
18 KiB
Java

// plasmaSwitchboardQueueEntry.java
// --------------------------------
// part of YaCy
// (C) by Michael Peter Christen; mc@anomic.de
// first published on http://www.anomic.de
// Frankfurt, Germany, 2005
//
// $LastChangedDate$
// $LastChangedRevision$
// $LastChangedBy$
//
// 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
//
// Using this software in any meaning (reading, learning, copying, compiling,
// running) means that you agree that the Author(s) is (are) not responsible
// for cost, loss of data or any harm that may be caused directly or indirectly
// by usage of this softare or this documentation. The usage of this software
// is on your own risk. The installation and usage (starting/running) of this
// software may allow other people or application to access your computer and
// any attached devices and is highly dependent on the configuration of the
// software which must be done by the user of the software; the author(s) is
// (are) also not responsible for proper configuration and usage of the
// software, even if provoked by documentation provided together with
// the software.
//
// Any changes to this file according to the GPL as documented in the file
// gpl.txt aside this file in the shipment you received can be done to the
// lines that follows this copyright notice here, but changes must not be
// done inside the copyright notive above. A re-distribution must contain
// the intact and unchanged copyright notice.
// Contributions and changes to the program code must be marked as such.
package de.anomic.plasma;
import java.io.File;
import java.io.IOException;
import java.net.MalformedURLException;
import java.util.Date;
import java.util.Iterator;
import de.anomic.index.indexURLEntry;
import de.anomic.kelondro.kelondroBase64Order;
import de.anomic.kelondro.kelondroNaturalOrder;
import de.anomic.kelondro.kelondroRow;
import de.anomic.kelondro.kelondroStack;
import de.anomic.plasma.cache.IResourceInfo;
import de.anomic.server.logging.serverLog;
import de.anomic.yacy.yacySeedDB;
import de.anomic.yacy.yacyURL;
public class plasmaSwitchboardQueue {
private kelondroStack sbQueueStack;
private plasmaCrawlProfile profiles;
private plasmaCrawlLURL lurls;
private File sbQueueStackPath;
public plasmaSwitchboardQueue(plasmaCrawlLURL lurls, File sbQueueStackPath, plasmaCrawlProfile profiles) {
this.sbQueueStackPath = sbQueueStackPath;
this.profiles = profiles;
this.lurls = lurls;
initQueueStack();
}
public static final kelondroRow rowdef = new kelondroRow(
"String url-256, " + // the url
"String refhash-" + yacySeedDB.commonHashLength + ", " + // the url's referrer hash
"Cardinal modifiedsince-11 {b64e}, " + // from ifModifiedSince
"byte[] flags-1, " + // flags
"String initiator-" + yacySeedDB.commonHashLength + ", " + // the crawling initiator
"Cardinal depth-2 {b64e}, " + // the prefetch depth so far, starts at 0
"String profile-" + yacySeedDB.commonHashLength + ", " + // the name of the prefetch profile handle
"String urldescr-80",
kelondroNaturalOrder.naturalOrder,
0);
private void initQueueStack() {
sbQueueStack = kelondroStack.open(sbQueueStackPath, rowdef);
}
/*
private void resetQueueStack() {
try {sbQueueStack.close();} catch (Exception e) {}
if (sbQueueStackPath.exists()) sbQueueStackPath.delete();
initQueueStack();
}
*/
public int size() {
return sbQueueStack.size();
}
public synchronized void push(Entry entry) throws IOException {
if (entry == null) return;
sbQueueStack.push(sbQueueStack.row().newEntry(new byte[][]{
entry.url.toString().getBytes(),
(entry.referrerHash == null) ? yacyURL.dummyHash.getBytes() : entry.referrerHash.getBytes(),
kelondroBase64Order.enhancedCoder.encodeLong((entry.ifModifiedSince == null) ? 0 : entry.ifModifiedSince.getTime(), 11).getBytes(),
new byte[]{entry.flags},
(entry.initiator == null) ? yacyURL.dummyHash.getBytes() : entry.initiator.getBytes(),
kelondroBase64Order.enhancedCoder.encodeLong((long) entry.depth, rowdef.width(5)).getBytes(),
(entry.profileHandle == null) ? yacyURL.dummyHash.getBytes() : entry.profileHandle.getBytes(),
(entry.anchorName == null) ? "-".getBytes("UTF-8") : entry.anchorName.getBytes("UTF-8")
}));
}
public synchronized Entry pop() throws IOException {
if (sbQueueStack.size() == 0) return null;
kelondroRow.Entry b = sbQueueStack.pot();
if (b == null) return null;
return new Entry(b);
}
public synchronized Entry remove(String urlHash) {
Iterator i = sbQueueStack.stackIterator(true);
kelondroRow.Entry rowentry;
Entry entry;
while (i.hasNext()) {
rowentry = (kelondroRow.Entry) i.next();
entry = new Entry(rowentry);
if (entry.urlHash().equals(urlHash)) {
i.remove();
return entry;
}
}
return null;
}
public void clear() {
sbQueueStack = kelondroStack.reset(sbQueueStack);
}
public void close() {
if (sbQueueStack != null) {
sbQueueStack.close();
}
sbQueueStack = null;
}
protected void finalize() throws Throwable {
try {
close();
} catch (Exception e) {
throw new IOException("plasmaSwitchboardQueue.finalize()" + e.getMessage());
}
super.finalize();
}
public Iterator entryIterator(boolean up) {
// iterates the elements in an ordered way.
// returns plasmaSwitchboardQueue.Entry - type Objects
return new entryIterator(up);
}
public class entryIterator implements Iterator {
Iterator rows;
public entryIterator(boolean up) {
rows = sbQueueStack.stackIterator(up);
}
public boolean hasNext() {
return rows.hasNext();
}
public Object next() {
return new Entry((kelondroRow.Entry) rows.next());
}
public void remove() {
rows.remove();
}
}
public Entry newEntry(yacyURL url, String referrer, Date ifModifiedSince, boolean requestWithCookie,
String initiator, int depth, String profilehandle, String anchorName) {
return new Entry(url, referrer, ifModifiedSince, requestWithCookie, initiator, depth, profilehandle, anchorName);
}
public class Entry {
private yacyURL url; // plasmaURL.urlStringLength
private String referrerHash; // plasmaURL.urlHashLength
private Date ifModifiedSince; // 6
private byte flags; // 1
private String initiator; // yacySeedDB.commonHashLength
private int depth; // plasmaURL.urlCrawlDepthLength
private String profileHandle; // plasmaURL.urlCrawlProfileHandleLength
private String anchorName; // plasmaURL.urlDescrLength
// computed values
private plasmaCrawlProfile.entry profileEntry;
private IResourceInfo contentInfo;
private yacyURL referrerURL;
public Entry(yacyURL url, String referrer, Date ifModifiedSince, boolean requestWithCookie,
String initiator, int depth, String profileHandle, String anchorName) {
this.url = url;
this.referrerHash = referrer;
this.ifModifiedSince = ifModifiedSince;
this.flags = (requestWithCookie) ? (byte) 1 : (byte) 0;
this.initiator = initiator;
this.depth = depth;
this.profileHandle = profileHandle;
this.anchorName = (anchorName==null)?"":anchorName.trim();
this.profileEntry = null;
this.contentInfo = null;
this.referrerURL = null;
}
public Entry(kelondroRow.Entry row) {
long ims = row.getColLong(2);
byte flags = row.getColByte(3);
try {
this.url = new yacyURL(row.getColString(0, "UTF-8"), null);
} catch (MalformedURLException e) {
this.url = null;
}
this.referrerHash = row.getColString(1, "UTF-8");
this.ifModifiedSince = (ims == 0) ? null : new Date(ims);
this.flags = ((flags & 1) == 1) ? (byte) 1 : (byte) 0;
this.initiator = row.getColString(4, "UTF-8");
this.depth = (int) row.getColLong(5);
this.profileHandle = row.getColString(6, "UTF-8");
this.anchorName = row.getColString(7, "UTF-8");
this.profileEntry = null;
this.contentInfo = null;
this.referrerURL = null;
}
public Entry(byte[][] row) throws IOException {
long ims = (row[2] == null) ? 0 : kelondroBase64Order.enhancedCoder.decodeLong(new String(row[2], "UTF-8"));
byte flags = (row[3] == null) ? 0 : row[3][0];
try {
this.url = new yacyURL(new String(row[0], "UTF-8"), null);
} catch (MalformedURLException e) {
this.url = null;
}
this.referrerHash = (row[1] == null) ? null : new String(row[1], "UTF-8");
this.ifModifiedSince = (ims == 0) ? null : new Date(ims);
this.flags = ((flags & 1) == 1) ? (byte) 1 : (byte) 0;
this.initiator = (row[4] == null) ? null : new String(row[4], "UTF-8");
this.depth = (int) kelondroBase64Order.enhancedCoder.decodeLong(new String(row[5], "UTF-8"));
this.profileHandle = new String(row[6], "UTF-8");
this.anchorName = (row[7] == null) ? null : (new String(row[7], "UTF-8")).trim();
this.profileEntry = null;
this.contentInfo = null;
this.referrerURL = null;
}
public yacyURL url() {
return url;
}
public String urlHash() {
return url.hash();
}
public boolean requestedWithCookie() {
return (flags & 1) == 1;
}
public File cacheFile() {
return plasmaHTCache.getCachePath(url);
}
public boolean proxy() {
return (initiator == null) || (initiator.equals(yacyURL.dummyHash));
}
public String initiator() {
return initiator;
}
public int depth() {
return depth;
}
public long size() {
if (cacheFile().exists()) return cacheFile().length(); else return 0;
}
public plasmaCrawlProfile.entry profile() {
if (profileEntry == null) profileEntry = profiles.getEntry(profileHandle);
return profileEntry;
}
private IResourceInfo getCachedObjectInfo() {
if (this.contentInfo == null) try {
this.contentInfo = plasmaHTCache.loadResourceInfo(this.url);
} catch (Exception e) {
serverLog.logSevere("PLASMA", "responseHeader: failed to get header", e);
return null;
}
return this.contentInfo;
}
public String getMimeType() {
IResourceInfo info = this.getCachedObjectInfo();
return (info == null) ? null : info.getMimeType();
}
public String getCharacterEncoding() {
IResourceInfo info = this.getCachedObjectInfo();
return (info == null) ? null : info.getCharacterEncoding();
}
public Date getModificationDate() {
IResourceInfo info = this.getCachedObjectInfo();
return (info == null) ? new Date() : info.getModificationDate();
}
public yacyURL referrerURL() {
if (referrerURL == null) {
if ((referrerHash == null) || (referrerHash.equals(yacyURL.dummyHash))) return null;
indexURLEntry entry = lurls.load(referrerHash, null);
if (entry == null) referrerURL = null; else referrerURL = entry.comp().url();
}
return referrerURL;
}
public String referrerHash() {
return referrerHash;
}
public String anchorName() {
return anchorName;
}
/**
* decide upon header information if a specific file should be indexed
* this method returns null if the answer is 'YES'!
* if the answer is 'NO' (do not index), it returns a string with the reason
* to reject the crawling demand in clear text
*
* This function is used by plasmaSwitchboard#processResourceStack
*/
public final String shallIndexCacheForProxy() {
if (profile() == null) {
return "shallIndexCacheForProxy: profile() is null !";
}
// check profile
if (!profile().indexText() && !profile().indexMedia()) {
return "Indexing_Not_Allowed";
}
// -CGI access in request
// CGI access makes the page very individual, and therefore not usable in caches
if (!profile().crawlingQ()) {
if (url.isPOST()) {
return "Dynamic_(POST)";
}
if (url.isCGI()) {
return "Dynamic_(CGI)";
}
}
// -authorization cases in request
// we checked that in shallStoreCache
// -ranges in request
// we checked that in shallStoreCache
// a picture cannot be indexed
if (plasmaHTCache.noIndexingURL(url)) {
return "Media_Content_(forbidden)";
}
// -cookies in request
// unfortunately, we cannot index pages which have been requested with a cookie
// because the returned content may be special for the client
if (requestedWithCookie()) {
// System.out.println("***not indexed because cookie");
return "Dynamic_(Requested_With_Cookie)";
}
if (getCachedObjectInfo() != null) {
return this.getCachedObjectInfo().shallIndexCacheForProxy();
}
return null;
}
/**
* decide upon header information if a specific file should be indexed
* this method returns null if the answer is 'YES'!
* if the answer is 'NO' (do not index), it returns a string with the reason
* to reject the crawling demand in clear text
*
* This function is used by plasmaSwitchboard#processResourceStack
*/
public final String shallIndexCacheForCrawler() {
if (profile() == null) {
return "shallIndexCacheForCrawler: profile() is null !";
}
// check profile
if (!profile().indexText() && !profile().indexMedia()) {
return "Indexing_Not_Allowed";
}
// -CGI access in request
// CGI access makes the page very individual, and therefore not usable in caches
if (!profile().crawlingQ()) {
if (url().isPOST()) { return "Dynamic_(POST)"; }
if (url().isCGI()) { return "Dynamic_(CGI)"; }
}
// -authorization cases in request
// we checked that in shallStoreCache
// -ranges in request
// we checked that in shallStoreCache
// a picture cannot be indexed
if (getCachedObjectInfo() != null) {
String status = this.getCachedObjectInfo().shallIndexCacheForCrawler();
if (status != null) return status;
}
if (plasmaHTCache.noIndexingURL(url())) { return "Media_Content_(forbidden)"; }
// -if-modified-since in request
// if the page is fresh at the very moment we can index it
// -> this does not apply for the crawler
// -cookies in request
// unfortunately, we cannot index pages which have been requested with a cookie
// because the returned content may be special for the client
// -> this does not apply for a crawler
// -set-cookie in response
// the set-cookie from the server does not indicate that the content is special
// thus we do not care about it here for indexing
// -> this does not apply for a crawler
// -pragma in cached response
// -> in the crawler we ignore this
// look for freshnes information
// -expires in cached response
// the expires value gives us a very easy hint when the cache is stale
// sometimes, the expires date is set to the past to prevent that a page is cached
// we use that information to see if we should index it
// -> this does not apply for a crawler
// -lastModified in cached response
// this information is too weak to use it to prevent indexing
// even if we can apply a TTL heuristic for cache usage
// -cache-control in cached response
// the cache-control has many value options.
// -> in the crawler we ignore this
return null;
}
} // class Entry
}