*) First trial implementation of robots.txt support

git-svn-id: https://svn.berlios.de/svnroot/repos/yacy/trunk@674 6c8d7289-2bf4-0310-a012-ef5d649a1542
pull/1/head
theli 20 years ago
parent 9444852896
commit f8ad65eae1

@ -6,7 +6,8 @@
// Frankfurt, Germany, 2004
//
// This file ist contributed by Alexander Schier
// last major change: 05.09.2005
// last major change: $LastChangedDate$ by $LastChangedBy$
// Revision: $LastChangedRevision$
//
// 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
@ -44,100 +45,227 @@
package de.anomic.data;
import java.lang.String;
import java.util.Vector;
import java.util.Iterator;
import java.io.IOException;
import java.net.MalformedURLException;
import java.net.URL;
import java.util.ArrayList;
import java.util.Date;
import java.util.HashSet;
import java.io.ByteArrayInputStream;
import java.io.ByteArrayOutputStream;
import java.io.File;
import java.io.BufferedReader;
import java.io.FileNotFoundException;
import java.io.FileReader;
import java.net.URL;
import java.io.IOException;
import java.io.InputStreamReader;
import de.anomic.http.httpHeader;
import de.anomic.http.httpc;
import de.anomic.plasma.plasmaCrawlRobotsTxt;
import de.anomic.plasma.plasmaParser;
import de.anomic.plasma.plasmaSwitchboard;
import de.anomic.plasma.plasmaURL;
import de.anomic.server.logging.serverLog;
import de.anomic.tools.bitfield;
import de.anomic.yacy.yacyCore;
/*
* A class for Parsing robots.txt files.
* It only parses the Deny Part, yet.
* TODO: Allow, Do not deny if User-Agent: != yacy
*
* http://www.robotstxt.org/wc/norobots-rfc.html
*
* Use:
* robotsParser rp=new Robotsparser(robotsfile);
* if(rp.isAllowedRobots("/test")){
* System.out.println("/test is allowed");
* }
*/
public class robotsParser{
String robotstxt="";
Vector deny;
/*
* constructor with a robots.txt as string
*/
public robotsParser(String robots){
robotstxt=robots;
parse();
}
/*
* constructor with a robots.txt as Filehandle
*/
public robotsParser(File robotsFile){
String robots="";
String line="";
try{
BufferedReader br=new BufferedReader(new FileReader(robotsFile));
while( (line=br.readLine()) != null){
robots+=line+"\n";
}
}catch(IOException e){
System.out.println("File Error");
}
robotstxt=robots;
parse();
}
public final class robotsParser{
/*public robotsParser(URL robotsUrl){
}*/
/*
* this parses the robots.txt.
* at the Moment it only creates a list of Deny Paths
*/
private void parse(){
deny=new Vector();
String[] lines = robotstxt.split("\n");
String line;
for(int i=0;i<lines.length;i++){
line=lines[i];
if(line.startsWith("Disallow:")){
line=line.substring(9);
line=line.trim();
deny.add(line);
}
}
}
/*
* Check if a url is allowed.
*/
public boolean isAllowedRobots(String url){
boolean allowed=false;
if(deny !=null){
Iterator it=deny.iterator();
allowed=true;
while(it.hasNext()){
if(url.startsWith((String)it.next())){
allowed=false;
break;
}
}
}
return allowed;
public static ArrayList parse(File robotsFile) throws IOException {
BufferedReader reader = new BufferedReader(new FileReader(robotsFile));
return parse(reader);
}
public static ArrayList parse(byte[] robotsTxt) throws IOException {
if ((robotsTxt == null)||(robotsTxt.length == 0)) return new ArrayList(0);
ByteArrayInputStream bin = new ByteArrayInputStream(robotsTxt);
BufferedReader reader = new BufferedReader(new InputStreamReader(bin));
return parse(reader);
}
public static ArrayList parse(BufferedReader reader) throws IOException{
ArrayList deny = new ArrayList();
int pos;
String line = null;
boolean rule4Yacy = false;
while ((line = reader.readLine()) != null) {
line = line.trim();
if (line.length() == 0) {
// we have reached the end of the rule block
rule4Yacy = false;
} else if (line.startsWith("#")) {
// we can ignore this. Just a comment line
} else if ((!rule4Yacy) && (line.startsWith("User-agent:"))) {
pos = line.indexOf(" ");
if (pos != -1) {
String userAgent = line.substring(pos).trim();
rule4Yacy = (userAgent.equals("*") || (userAgent.toLowerCase().indexOf("yacy") >=0));
}
} else if (line.startsWith("Disallow:") && rule4Yacy) {
pos = line.indexOf(" ");
if (pos != -1) {
// getting the path
String path = line.substring(pos).trim();
// escaping all occurences of ; because this char is used as special char in the Robots DB
path = path.replaceAll(";","%3B");
// adding it to the pathlist
deny.add(path);
}
}
}
return deny;
}
/*
* Test the class with a file robots.txt in the workdir and a testpath as argument.
*/
public static void main(String[] args){
robotsParser rp=new robotsParser(new File("robots.txt"));
for(int i=0;i<args.length;i++){
if(rp.isAllowedRobots(args[i])){
System.out.println(args[i]+" is allowed.");
}else{
System.out.println(args[i]+" is NOT allowed.");
}
}
}
public static boolean isDisallowed(URL nexturl) {
if (nexturl == null) throw new IllegalArgumentException();
// generating the hostname:poart string needed to do a DB lookup
String urlHostPort = nexturl.getHost() + ":" + ((nexturl.getPort()==-1)?80:nexturl.getPort());
// doing a DB lookup to determine if the robots data is already available
plasmaCrawlRobotsTxt.Entry robotsTxt4Host = plasmaSwitchboard.robots.getEntry(urlHostPort);
// if we have not found any data or the data is older than 7 days, we need to load it from the remote server
if ((robotsTxt4Host == null) ||
(System.currentTimeMillis() - robotsTxt4Host.getLoadedDate().getTime() > 7*24*60*60*1000)) {
URL robotsURL = null;
// generating the proper url to download the robots txt
try {
robotsURL = new URL(nexturl.getProtocol(),nexturl.getHost(),(nexturl.getPort()==-1)?80:nexturl.getPort(),"/robots.txt");
} catch (MalformedURLException e) {
serverLog.logSevere("ROBOTS","Unable to generate robots.txt URL for URL '" + nexturl.toString() + "'.");
return false;
}
boolean accessCompletelyRestricted = false;
byte[] robotsTxt = null;
try {
Object[] result = downloadRobotsTxt(robotsURL,5);
accessCompletelyRestricted = ((Boolean)result[0]).booleanValue();
robotsTxt = (byte[])result[1];
} catch (Exception e) {
serverLog.logSevere("ROBOTS","Unable to download the robots.txt file from URL '" + robotsURL + "'. " + e.getMessage());
}
ArrayList denyPath = null;
if (accessCompletelyRestricted) {
denyPath = new ArrayList();
denyPath.add("/");
} else {
// parsing the robots.txt Data and converting it into an arraylist
try {
denyPath = robotsParser.parse(robotsTxt);
} catch (IOException e) {
serverLog.logSevere("ROBOTS","Unable to parse the robots.txt file from URL '" + robotsURL + "'.");
}
}
// storing the data into the robots DB
robotsTxt4Host = plasmaSwitchboard.robots.addEntry(urlHostPort,denyPath,new Date());
}
if (robotsTxt4Host.isDisallowed(nexturl.getPath())) {
return true;
}
return false;
}
private static Object[] downloadRobotsTxt(URL robotsURL, int redirectionCount) throws Exception {
if (redirectionCount < 0) return new Object[]{Boolean.FALSE,null};
redirectionCount--;
boolean accessCompletelyRestricted = false;
byte[] robotsTxt = null;
httpc con = null;
try {
plasmaSwitchboard sb = plasmaSwitchboard.getSwitchboard();
if (!sb.remoteProxyUse) {
con = httpc.getInstance(robotsURL.getHost(), robotsURL.getPort(), 10000, false);
} else {
con = httpc.getInstance(robotsURL.getHost(), robotsURL.getPort(), 10000, false, sb.remoteProxyHost, sb.remoteProxyPort);
}
httpc.response res = con.GET(robotsURL.getPath(), null);
if (res.status.startsWith("2")) {
if (!res.responseHeader.mime().startsWith("text/plain")) {
robotsTxt = null;
serverLog.logFinest("ROBOTS","Robots.txt from URL '" + robotsURL + "' has wrong mimetype '" + res.responseHeader.mime() + "'.");
} else {
ByteArrayOutputStream bos = new ByteArrayOutputStream();
res.writeContent(bos, null);
con.close();
robotsTxt = bos.toByteArray();
serverLog.logFinest("ROBOTS","Robots.txt successfully loaded from URL '" + robotsURL + "'.");
}
} else if (res.status.startsWith("3")) {
// getting redirection URL
String redirectionUrlString = (String) res.responseHeader.get(httpHeader.LOCATION);
redirectionUrlString = redirectionUrlString.trim();
// generating the new URL object
URL redirectionUrl = new URL(robotsURL, redirectionUrlString);
// returning the used httpc
httpc.returnInstance(con);
con = null;
// following the redirection
serverLog.logFinest("ROBOTS","Redirection detected for Robots.txt with URL '" + robotsURL + "'." +
"\nRedirecting request to: " + redirectionUrl);
return downloadRobotsTxt(redirectionUrl,redirectionCount);
} else if (res.status.startsWith("401") || res.status.startsWith("403")) {
accessCompletelyRestricted = true;
serverLog.logFinest("ROBOTS","Access to Robots.txt not allowed on URL '" + robotsURL + "'.");
} else {
serverLog.logFinest("ROBOTS","Robots.txt could not be downloaded from URL '" + robotsURL + "'. [" + res.status + "].");
robotsTxt = null;
}
} catch (Exception e) {
throw e;
} finally {
if (con != null) httpc.returnInstance(con);
}
return new Object[]{new Boolean(accessCompletelyRestricted),robotsTxt};
}
// /*
// * Test the class with a file robots.txt in the workdir and a testpath as argument.
// */
// public static void main(String[] args){
// robotsParser rp=new robotsParser(new File("robots.txt"));
// for(int i=0;i<args.length;i++){
// if(rp.isAllowedRobots(args[i])){
// System.out.println(args[i]+" is allowed.");
// }else{
// System.out.println(args[i]+" is NOT allowed.");
// }
// }
// }
}

@ -1019,6 +1019,19 @@ public final class httpdProxyHandler extends httpdAbstractHandler implements htt
String httpVersion = conProp.getProperty("HTTP");
int timeout = Integer.parseInt(switchboard.getConfig("clientTimeout", "10000"));
// check the blacklist
// blacklist idea inspired by [AS]:
// respond a 404 for all AGIS ("all you get is shit") servers
String hostlow = host.toLowerCase();
if (plasmaSwitchboard.urlBlacklist.isListed(hostlow, "/")) {
httpd.sendRespondError(conProp,clientOut,4,403,null,
"URL '" + hostlow + "' blocked by yacy proxy (blacklisted)",null);
this.theLogger.logInfo("AGIS blocking of host '" + hostlow + "'");
forceConnectionClose();
return;
}
// possibly branch into PROXY-PROXY connection
if (remoteProxyUse) {
httpc remoteProxy = null;
@ -1035,6 +1048,7 @@ public final class httpdProxyHandler extends httpdAbstractHandler implements htt
// pass error response back to client
httpd.sendRespondHeader(conProp,clientOut,httpVersion,response.statusCode,response.statusText,response.responseHeader);
//respondHeader(clientOut, response.status, response.responseHeader);
forceConnectionClose();
return;
}
} catch (Exception e) {

@ -0,0 +1,213 @@
//plasmaCrawlRobotsTxt.java
//-------------------------------------
//part of YACY
//(C) by Michael Peter Christen; mc@anomic.de
//first published on http://www.anomic.de
//Frankfurt, Germany, 2004
//
//This file ist contributed by Martin Thelian
//last major change: $LastChangedDate$ by $LastChangedBy$
//Revision: $LastChangedRevision$
//
//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.util.ArrayList;
import java.util.Arrays;
import java.util.Date;
import java.util.HashMap;
import java.util.Iterator;
import java.util.LinkedList;
import java.util.Map;
import de.anomic.kelondro.kelondroDyn;
import de.anomic.kelondro.kelondroMap;
import de.anomic.server.logging.serverLog;
public class plasmaCrawlRobotsTxt {
private kelondroMap robotsTable;
private File robotsTableFile;
public plasmaCrawlRobotsTxt(File robotsTableFile) throws IOException {
this.robotsTableFile = robotsTableFile;
if (robotsTableFile.exists()) {
robotsTable = new kelondroMap(new kelondroDyn(robotsTableFile, 32000));
} else {
robotsTableFile.getParentFile().mkdirs();
robotsTable = new kelondroMap(new kelondroDyn(robotsTableFile, 32000, 256, 512));
}
}
private void resetDatabase() {
// deletes the robots.txt database and creates a new one
if (robotsTable != null) try {
robotsTable.close();
} catch (IOException e) {}
if (!(robotsTableFile.delete())) throw new RuntimeException("cannot delete robots.txt database");
try {
robotsTableFile.getParentFile().mkdirs();
robotsTable = new kelondroMap(new kelondroDyn(robotsTableFile, 32000, 256, 512));
} catch (IOException e){
serverLog.logSevere("PLASMA", "robotsTxt.resetDatabase", e);
}
}
public void close() {
try {
robotsTable.close();
} catch (IOException e) {}
}
public int size() {
return robotsTable.size();
}
public void removeEntry(String hostName) {
try {
robotsTable.remove(hostName.toLowerCase());
} catch (IOException e) {}
}
public Entry getEntry(String hostName) {
try {
Map record = robotsTable.get(hostName);
if (record == null) return null;
return new Entry(hostName, record);
} catch (IOException e) {
return null;
}
}
public Entry addEntry(String hostName, ArrayList disallowPathList, Date loadedDate) {
Entry entry = new Entry(hostName,disallowPathList,loadedDate);
addEntry(entry);
return entry;
}
public String addEntry(Entry entry) {
// writes a new page and returns key
try {
robotsTable.set(entry.hostName,entry.mem);
return entry.hostName;
} catch (IOException e) {
return null;
}
}
public class Entry {
public static final String DISALLOW_PATH_LIST = "disallow";
public static final String LOADED_DATE = "date";
// this is a simple record structure that hold all properties of a single crawl start
private Map mem;
private LinkedList disallowPathList;
private String hostName;
public Entry(String hostName, Map mem) {
this.hostName = hostName;
this.mem = mem;
if (this.mem.containsKey(DISALLOW_PATH_LIST)) {
this.disallowPathList = new LinkedList();
String csPl = (String) this.mem.get(DISALLOW_PATH_LIST);
if (csPl.length() > 0){
String[] pathArray = csPl.split(";");
if ((pathArray != null)&&(pathArray.length > 0)) {
this.disallowPathList.addAll(Arrays.asList(pathArray));
}
}
} else {
this.disallowPathList = new LinkedList();
}
}
public Entry(String hostName, ArrayList disallowPathList, Date loadedDate) {
if ((hostName == null) || (hostName.length() == 0)) throw new IllegalArgumentException();
this.hostName = hostName.trim().toLowerCase();
this.disallowPathList = new LinkedList();
this.mem = new HashMap();
if (loadedDate != null) this.mem.put(LOADED_DATE,Long.toString(loadedDate.getTime()));
if ((disallowPathList != null)&&(disallowPathList.size()>0)) {
this.disallowPathList.addAll(disallowPathList);
StringBuffer pathListStr = new StringBuffer();
for (int i=0; i<disallowPathList.size();i++) {
pathListStr.append(disallowPathList.get(i))
.append(";");
}
this.mem.put(DISALLOW_PATH_LIST,pathListStr.substring(0,pathListStr.length()-1));
}
}
public String toString() {
StringBuffer str = new StringBuffer();
str.append((this.hostName==null)?"null":this.hostName)
.append(": ");
if (this.mem != null) {
str.append(this.mem.toString());
}
return str.toString();
}
public Date getLoadedDate() {
if (this.mem.containsKey(LOADED_DATE)) {
return new Date(Long.valueOf((String) this.mem.get(LOADED_DATE)).longValue());
}
return null;
}
public boolean isDisallowed(String path) {
if ((path == null) || (path.length() == 0)) return false;
if ((this.mem == null) || (this.disallowPathList.size() == 0)) return false;
Iterator pathIter = this.disallowPathList.iterator();
while (pathIter.hasNext()) {
String nextPath = (String) pathIter.next();
if (path.startsWith(nextPath)) return true;
}
return false;
}
}
}

@ -4,7 +4,9 @@
// (C) by Michael Peter Christen; mc@anomic.de
// first published on http://www.anomic.de
// Frankfurt, Germany, 2004, 2005
// last major change: 24.03.2005
//
// last major change: $LastChangedDate$ by $LastChangedBy$
// Revision: $LastChangedRevision$
//
// 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
@ -106,6 +108,7 @@ import java.io.IOException;
import java.net.MalformedURLException;
import java.net.URL;
import java.text.SimpleDateFormat;
import java.util.ArrayList;
import java.util.Date;
import java.util.HashMap;
import java.util.HashSet;
@ -114,6 +117,7 @@ import java.util.Map;
import java.util.Set;
import java.util.TreeSet;
import de.anomic.data.messageBoard;
import de.anomic.data.robotsParser;
import de.anomic.data.wikiBoard;
import de.anomic.htmlFilter.htmlFilterContentScraper;
import de.anomic.http.httpHeader;
@ -163,6 +167,7 @@ public final class plasmaSwitchboard extends serverAbstractSwitch implements ser
public String remoteProxyHost;
public int remoteProxyPort;
public boolean remoteProxyUse;
public static plasmaCrawlRobotsTxt robots;
public plasmaCrawlProfile profiles;
public plasmaCrawlProfile.entry defaultProxyProfile;
public plasmaCrawlProfile.entry defaultRemoteProfile;
@ -259,10 +264,16 @@ public final class plasmaSwitchboard extends serverAbstractSwitch implements ser
// make crawl profiles database and default profiles
this.log.logConfig("Initializing Crawl Profiles");
File profilesFile = new File(this.plasmaPath, "crawlProfiles0.db");
this.profiles = new plasmaCrawlProfile(new File(this.plasmaPath, "crawlProfiles0.db"));
this.profiles = new plasmaCrawlProfile(profilesFile);
initProfiles();
log.logConfig("Loaded profiles from file " + profilesFile + ", " + this.profiles.size() + " entries");
// loading the robots.txt db
this.log.logConfig("Initializing robots.txt DB");
File robotsDBFile = new File(this.plasmaPath, "crawlRobotsTxt.db");
this.robots = new plasmaCrawlRobotsTxt(robotsDBFile);
this.log.logConfig("Loaded robots.txt DB from file " + robotsDBFile + ", " + this.robots.size() + " entries");
// start indexing management
log.logConfig("Starting Indexing Management");
urlPool = new plasmaURLPool(plasmaPath, ramLURL, ramNURL, ramEURL);
@ -572,6 +583,7 @@ public final class plasmaSwitchboard extends serverAbstractSwitch implements ser
if (facilityDB != null) facilityDB.close();
urlPool.close();
profiles.close();
robots.close();
parser.close();
cacheManager.close();
sbQueue.close();
@ -1148,6 +1160,16 @@ public final class plasmaSwitchboard extends serverAbstractSwitch implements ser
log.logFine("URL '" + nexturlString + "' is post URL.");
return reason;
}
// checking robots.txt
if (robotsParser.isDisallowed(nexturl)) {
reason = "denied_(robots.txt)";
urlPool.errorURL.newEntry(nexturl, referrerHash, initiatorHash, yacyCore.seedDB.mySeed.hash,
name, reason, new bitfield(plasmaURL.urlFlagLength), false);
log.logFine("Crawling of URL '" + nexturlString + "' disallowed by robots.txt.");
return reason;
}
String nexturlhash = plasmaURL.urlHash(nexturl);
String dbocc = "";

Loading…
Cancel
Save