yacy_search_server/htroot/yacy/search.java

341 lines
20 KiB
Java
Raw Normal View History

// search.java
// (C) 2004 by Michael Peter Christen; mc@yacy.net, Frankfurt a. M., Germany
// first published 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
// You must compile this file with
// javac -classpath .:../../Classes search.java
// if the shell's current path is htroot/yacy
import java.util.HashMap;
import java.util.Iterator;
import java.util.Map;
import java.util.TreeSet;
import de.anomic.http.httpHeader;
import de.anomic.kelondro.kelondroBase64Order;
import de.anomic.kelondro.kelondroBitfield;
import de.anomic.index.indexContainer;
import de.anomic.net.natLib;
import de.anomic.plasma.plasmaURL;
import de.anomic.index.indexURLEntry;
import de.anomic.plasma.plasmaCondenser;
import de.anomic.plasma.plasmaSearchEvent;
import de.anomic.plasma.plasmaSearchPreOrder;
import de.anomic.plasma.plasmaSearchQuery;
import de.anomic.plasma.plasmaSearchRankingProfile;
import de.anomic.plasma.plasmaSearchPostOrder;
import de.anomic.plasma.plasmaSearchProcessing;
import de.anomic.plasma.plasmaSnippetCache;
import de.anomic.plasma.plasmaSwitchboard;
import de.anomic.plasma.plasmaWordIndex;
import de.anomic.server.serverCore;
import de.anomic.server.serverObjects;
import de.anomic.server.serverSwitch;
import de.anomic.yacy.yacyCore;
import de.anomic.yacy.yacyDHTAction;
import de.anomic.yacy.yacyNetwork;
import de.anomic.yacy.yacySeed;
import de.anomic.tools.crypt;
*) Asynchronous queuing of crawl job URLs (stackCrawl) various checks like the blacklist check or the robots.txt disallow check are now done by a separate thread to unburden the indexer thread(s) TODO: maybe we have to introduce a threadpool here if it turn out that this single thread is a bottleneck because of the time consuming robots.txt downloads *) improved index transfer The index selection and transmission is done in parallel now to improve index transfer performance. TODO: maybe we could speed up performance by unsing multiple transmission threads in parallel instead of only a single one. *) gzip encoded post requests it is now configureable if a gzip encoded post request should be send on intex transfer/distribution *) storage Peer (very experimentell and not optimized yet) Now it's possible to send the result of the yacy indexer thread to a remote peer istead of storing the indexed words locally. This could be done by setting the property "storagePeerHash" in the yacy config file - Please note that if the index transfer fails, the index ist stored locally. - TODO: currently this index transfer is done by the indexer thread. To seedup the indexer a) this transmission should be done in parallel and b) multiple chunks should be bundled and transfered together *) general performance improvements - better memory cleanup after http request processing has finished - replacing some string concatenations with stringBuffers - replacing BufferedInputStreams with serverByteBuffer - replacing vectors with arraylists wherever possible - replacing hashtables with hashmaps wherever possible This was done because function calls to verctor or hashtable functions take 3 time longer than calls to functions of arraylists or hashmaps. TODO: we should take a look on the class serverObject which is inherited from hashmap Do we realy need a synchronization for this class? TODO: replace arraylists with linkedLists if random access to the list elements is not needed *) Robots Parser supports if-modified-since downloads now If the downloaded robots.txt file is older than 7 days the robots parser tries to download the robots.txt with the if-modified-since header to avoid unnecessary downloads if the file was not changed. Additionally the ETag header is used to detect changes. *) Crawler: better handling of unsupported mimeTypes + FileExtension *) Bugfix: plasmaWordIndexEntity was not closed correctly in - query.java - plasmaswitchboard.java *) function minimizeUrlDB added to yacy.java this function tests the current urlHashDB for unused urls ATTENTION: please don't use this function at the moment because it causes the wordIndexDB to flush all words into the word directory! git-svn-id: https://svn.berlios.de/svnroot/repos/yacy/trunk@853 6c8d7289-2bf4-0310-a012-ef5d649a1542
2005-10-05 12:45:33 +02:00
public final class search {
public static serverObjects respond(httpHeader header, serverObjects post, serverSwitch env) {
// return variable that accumulates replacements
final plasmaSwitchboard sb = (plasmaSwitchboard) env;
serverObjects prop = new serverObjects();
if ((post == null) || (env == null)) return prop;
if (!yacyNetwork.authentifyRequest(post, env)) return prop;
//System.out.println("yacy: search received request = " + post.toString());
final String oseed = post.get("myseed", ""); // complete seed of the requesting peer
// final String youare = post.get("youare", ""); // seed hash of the target peer, used for testing network stability
final String key = post.get("key", ""); // transmission key for response
final String query = post.get("query", ""); // a string of word hashes that shall be searched and combined
final String exclude= post.get("exclude", "");// a string of word hashes that shall not be within the search result
String urls = post.get("urls", ""); // a string of url hashes that are preselected for the search: no other may be returned
String abstracts = post.get("abstracts", ""); // a string of word hashes for abstracts that shall be generated, or 'auto' (for maxcount-word), or '' (for none)
// final String fwdep = post.get("fwdep", ""); // forward depth. if "0" then peer may NOT ask another peer for more results
// final String fwden = post.get("fwden", ""); // forward deny, a list of seed hashes. They may NOT be target of forward hopping
final long duetime= Math.min(60000, post.getLong("duetime", 3000));
final int count = Math.min(100, post.getInt("count", 10)); // maximum number of wanted results
final int maxdist= post.getInt("maxdist", Integer.MAX_VALUE);
final String prefer = post.get("prefer", "");
final String contentdom = post.get("contentdom", "text");
final String filter = post.get("filter", ".*");
final int partitions = post.getInt("partitions", 30);
String profile = post.get("profile", ""); // remote profile hand-over
if (profile.length() > 0) profile = crypt.simpleDecode(profile, null);
final boolean includesnippet = post.get("includesnippet", "false").equals("true");
final kelondroBitfield constraint = new kelondroBitfield(4, post.get("constraint", "______"));
// final boolean global = ((String) post.get("resource", "global")).equals("global"); // if true, then result may consist of answers from other peers
// Date remoteTime = yacyCore.parseUniversalDate((String) post.get(yacySeed.MYTIME)); // read remote time
// test:
// http://localhost:8080/yacy/search.html?query=4galTpdpDM5Q (search for linux)
// http://localhost:8080/yacy/search.html?query=gh8DKIhGKXws (search for book)
// http://localhost:8080/yacy/search.html?query=4galTpdpDM5Qgh8DKIhGKXws&abstracts=auto (search for linux and book, generate abstract automatically)
// http://localhost:8080/yacy/search.html?query=&abstracts=4galTpdpDM5Q (only abstracts for linux)
if ((sb.isRobinsonMode()) &&
(!((sb.isPublicRobinson()) ||
(sb.isInMyCluster((String)header.get(httpHeader.CONNECTION_PROP_CLIENTIP)))))) {
// if we are a robinson cluster, answer only if this client is known by our network definition
prop.putASIS("links", "");
prop.putASIS("linkcount", "0");
prop.putASIS("references", "");
return prop;
}
// tell all threads to do nothing for a specific time
sb.intermissionAllThreads(2 * duetime);
TreeSet abstractSet = ((abstracts.length() == 0) || (abstracts.equals("auto"))) ? null : plasmaSearchQuery.hashes2Set(abstracts);
// store accessing peer
if (yacyCore.seedDB == null) {
yacyCore.log.logSevere("yacy.search: seed cache not initialized");
} else {
yacyCore.peerActions.peerArrival(yacySeed.genRemoteSeed(oseed, key, true), true);
}
// prepare search
final TreeSet queryhashes = plasmaSearchQuery.hashes2Set(query);
final TreeSet excludehashes = (exclude.length() == 0) ? new TreeSet(kelondroBase64Order.enhancedCoder) : plasmaSearchQuery.hashes2Set(exclude);
final long timestamp = System.currentTimeMillis();
// prepare an abstract result
StringBuffer indexabstract = new StringBuffer();
int indexabstractContainercount = 0;
int joincount = 0;
plasmaSearchPostOrder acc = null;
plasmaSearchQuery squery = null;
//plasmaSearchEvent theSearch = null;
if ((query.length() == 0) && (abstractSet != null)) {
// this is _not_ a normal search, only a request for index abstracts
squery = new plasmaSearchQuery(abstractSet, new TreeSet(kelondroBase64Order.enhancedCoder), maxdist, prefer, plasmaSearchQuery.contentdomParser(contentdom), count, duetime, filter, plasmaSearchQuery.catchall_constraint);
squery.domType = plasmaSearchQuery.SEARCHDOM_LOCAL;
yacyCore.log.logInfo("INIT HASH SEARCH (abstracts only): " + plasmaSearchQuery.anonymizedQueryHashes(squery.queryHashes) + " - " + squery.wantedResults + " links");
// prepare a search profile
//plasmaSearchRankingProfile rankingProfile = (profile.length() == 0) ? new plasmaSearchRankingProfile(contentdom) : new plasmaSearchRankingProfile("", profile);
plasmaSearchProcessing localTiming = new plasmaSearchProcessing(squery.maximumTime, squery.wantedResults);
//plasmaSearchProcessing remoteTiming = null;
//theSearch = new plasmaSearchEvent(squery, rankingProfile, localTiming, remoteTiming, true, sb.wordIndex, null);
Map[] containers = localTiming.localSearchContainers(squery, sb.wordIndex, plasmaSearchQuery.hashes2Set(urls));
if (containers != null) {
Iterator ci = containers[0].entrySet().iterator();
Map.Entry entry;
String wordhash;
while (ci.hasNext()) {
entry = (Map.Entry) ci.next();
wordhash = (String) entry.getKey();
indexContainer container = (indexContainer) entry.getValue();
indexabstractContainercount += container.size();
indexabstract.append("indexabstract." + wordhash + "=").append(plasmaURL.compressIndex(container, null, 1000).toString()).append(serverCore.crlfString);
}
}
prop.putASIS("indexcount", "");
prop.put("joincount", 0);
} else {
// retrieve index containers from search request
squery = new plasmaSearchQuery(queryhashes, excludehashes, maxdist, prefer, plasmaSearchQuery.contentdomParser(contentdom), count, duetime, filter, constraint);
squery.domType = plasmaSearchQuery.SEARCHDOM_LOCAL;
yacyCore.log.logInfo("INIT HASH SEARCH (query-" + abstracts + "): " + plasmaSearchQuery.anonymizedQueryHashes(squery.queryHashes) + " - " + squery.wantedResults + " links");
// prepare a search profile
plasmaSearchRankingProfile rankingProfile = (profile.length() == 0) ? new plasmaSearchRankingProfile(contentdom) : new plasmaSearchRankingProfile("", profile);
plasmaSearchProcessing localProcess = new plasmaSearchProcessing(squery.maximumTime, squery.wantedResults);
//plasmaSearchProcessing remoteProcess = null;
//theSearch = new plasmaSearchEvent(squery, rankingProfile, localProcess, remoteProcess, true, sb.wordIndex, null);
Map[] containers = localProcess.localSearchContainers(squery, sb.wordIndex, plasmaSearchQuery.hashes2Set(urls));
// set statistic details of search result and find best result index set
if (containers == null) {
prop.putASIS("indexcount", "");
prop.putASIS("joincount", "0");
} else {
Iterator ci = containers[0].entrySet().iterator();
StringBuffer indexcount = new StringBuffer();
Map.Entry entry;
int maxcount = -1;
double mindhtdistance = 1.1, d;
String wordhash;
String maxcounthash = null, neardhthash = null;
while (ci.hasNext()) {
entry = (Map.Entry) ci.next();
wordhash = (String) entry.getKey();
indexContainer container = (indexContainer) entry.getValue();
if (container.size() > maxcount) {
maxcounthash = wordhash;
maxcount = container.size();
}
d = yacyDHTAction.dhtDistance(yacyCore.seedDB.mySeed.hash, wordhash);
if (d < mindhtdistance) {
// calculate the word hash that is closest to our dht position
mindhtdistance = d;
neardhthash = wordhash;
}
indexcount.append("indexcount.").append(container.getWordHash()).append('=').append(Integer.toString(container.size())).append(serverCore.crlfString);
if ((abstractSet != null) && (abstractSet.contains(wordhash))) {
// if a specific index-abstract is demanded, attach it here
indexabstractContainercount += container.size();
indexabstract.append("indexabstract." + wordhash + "=").append(plasmaURL.compressIndex(container, null,1000).toString()).append(serverCore.crlfString);
}
}
prop.putASIS("indexcount", new String(indexcount));
// join and order the result
indexContainer localResults =
(containers == null) ?
plasmaWordIndex.emptyContainer(null) :
localProcess.localSearchJoinExclude(
containers[0].values(),
containers[1].values(),
(squery.queryHashes.size() == 0) ?
0 :
localProcess.getTargetTime(plasmaSearchProcessing.PROCESS_JOIN) * squery.queryHashes.size() / (squery.queryHashes.size() + squery.excludeHashes.size()),
squery.maxDistance);
if (localResults == null) {
joincount = 0;
prop.put("joincount", 0);
acc = null;
} else {
joincount = localResults.size();
prop.putASIS("joincount", Integer.toString(joincount));
plasmaSearchPreOrder pre = localProcess.preSort(squery, rankingProfile, localResults);
acc = localProcess.urlFetch(squery, rankingProfile, sb.wordIndex, pre);
acc.localContributions = (localResults == null) ? 0 : localResults.size();
localProcess.postSort(true, acc);
localProcess.applyFilter(acc);
}
// generate compressed index for maxcounthash
// this is not needed if the search is restricted to specific
// urls, because it is a re-search
if ((maxcounthash == null) || (urls.length() != 0) || (queryhashes.size() == 1) || (abstracts.length() == 0)) {
prop.putASIS("indexabstract", "");
} else if (abstracts.equals("auto")) {
// automatically attach the index abstract for the index that has the most references. This should be our target dht position
indexContainer container = (indexContainer) containers[0].get(maxcounthash);
indexabstractContainercount += container.size();
indexabstract.append("indexabstract." + maxcounthash + "=").append(plasmaURL.compressIndex(container,localResults, 1000).toString()).append(serverCore.crlfString);
if ((neardhthash != null) && (!(neardhthash.equals(maxcounthash)))) {
// in case that the neardhthash is different from the maxcounthash attach also the neardhthash-container
container = (indexContainer) containers[0].get(neardhthash);
indexabstractContainercount += container.size();
indexabstract.append("indexabstract." + neardhthash + "=").append(plasmaURL.compressIndex(container, localResults, 1000).toString()).append(serverCore.crlfString);
}
//System.out.println("DEBUG-ABSTRACTGENERATION: maxcounthash = " + maxcounthash);
//System.out.println("DEBUG-ABSTRACTGENERATION: neardhthash = "+ neardhthash);
//yacyCore.log.logFine("DEBUG HASH SEARCH: " + indexabstract);
}
}
if (partitions > 0) sb.requestedQueries = sb.requestedQueries + 1d / (double) partitions; // increase query counter
}
prop.putASIS("indexabstract", new String(indexabstract));
// prepare search statistics
Long trackerHandle = new Long(System.currentTimeMillis());
String client = (String) header.get("CLIENTIP");
HashMap searchProfile = plasmaSearchEvent.resultProfile(squery, joincount, System.currentTimeMillis() - timestamp);
searchProfile.put("host", client);
yacySeed remotepeer = yacyCore.seedDB.lookupByIP(natLib.getInetAddress(client), true, false, false);
searchProfile.put("peername", (remotepeer == null) ? "unknown" : remotepeer.getName());
searchProfile.put("time", trackerHandle);
sb.remoteSearches.add(searchProfile);
TreeSet handles = (TreeSet) sb.remoteSearchTracker.get(client);
if (handles == null) handles = new TreeSet();
handles.add(trackerHandle);
sb.remoteSearchTracker.put(client, handles);
// prepare result
if ((joincount == 0) || (acc == null)) {
// no results
prop.putASIS("links", "");
prop.putASIS("linkcount", "0");
prop.putASIS("references", "");
} else {
// result is a List of urlEntry elements
int i = 0;
StringBuffer links = new StringBuffer();
String resource = null;
indexURLEntry urlentry;
plasmaSnippetCache.TextSnippet snippet;
while ((acc.hasMoreElements()) && (i < squery.wantedResults)) {
urlentry = (indexURLEntry) acc.nextElement();
if (includesnippet) {
snippet = plasmaSnippetCache.retrieveTextSnippet(urlentry.comp().url(), squery.queryHashes, false, urlentry.flags().get(plasmaCondenser.flag_cat_indexof), 260, 1000);
} else {
snippet = null;
}
if ((snippet != null) && (snippet.exists())) {
resource = urlentry.toString(snippet.getLineRaw());
} else {
resource = urlentry.toString();
}
if (resource != null) {
links.append("resource").append(i).append('=').append(resource).append(serverCore.crlfString);
i++;
}
}
prop.putASIS("links", new String(links));
prop.putASIS("linkcount", Integer.toString(i));
// prepare reference hints
Object[] ws = acc.getReferences(16);
StringBuffer refstr = new StringBuffer();
for (int j = 0; j < ws.length; j++)
refstr.append(",").append((String) ws[j]);
prop.putASIS("references", (refstr.length() > 0) ? refstr.substring(1) : new String(refstr));
}
// add information about forward peers
prop.putASIS("fwhop", ""); // hops (depth) of forwards that had been performed to construct this result
prop.putASIS("fwsrc", ""); // peers that helped to construct this result
prop.putASIS("fwrec", ""); // peers that would have helped to construct this result (recommendations)
// log
yacyCore.log.logInfo("EXIT HASH SEARCH: " +
plasmaSearchQuery.anonymizedQueryHashes(squery.queryHashes) + " - " + joincount + " links found, " +
prop.get("linkcount", "?") + " links selected, " +
indexabstractContainercount + " index abstract references attached, " +
(System.currentTimeMillis() - timestamp) + " milliseconds");
prop.putASIS("searchtime", Long.toString(System.currentTimeMillis() - timestamp));
final int links = Integer.parseInt(prop.get("linkcount","0"));
yacyCore.seedDB.mySeed.incSI(links);
yacyCore.seedDB.mySeed.incSU(links);
return prop;
}
}