2005-09-07 13:17:21 +02:00
//plasmaCrawlRobotsTxt.java
//-------------------------------------
//part of YACY
2008-07-20 19:14:51 +02:00
//(C) by Michael Peter Christen; mc@yacy.net
2005-09-07 13:17:21 +02:00
//first published on http://www.anomic.de
//Frankfurt, Germany, 2004
//
2008-07-05 02:35:20 +02:00
//This file is contributed by Martin Thelian
// [MC] moved some methods from robotsParser file that had been created by Alexander Schier to this class
2005-09-07 13:17:21 +02:00
//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
2008-05-06 02:32:41 +02:00
package de.anomic.crawler ;
2005-09-07 13:17:21 +02:00
2008-07-05 02:35:20 +02:00
import java.io.BufferedInputStream ;
import java.io.BufferedOutputStream ;
2005-09-07 13:17:21 +02:00
import java.io.File ;
import java.io.IOException ;
2008-07-05 02:35:20 +02:00
import java.net.MalformedURLException ;
2005-09-07 13:17:21 +02:00
import java.util.ArrayList ;
import java.util.Arrays ;
import java.util.Date ;
import java.util.HashMap ;
import java.util.Iterator ;
import java.util.LinkedList ;
2008-10-10 10:39:11 +02:00
import java.util.Map ;
2008-07-11 14:03:18 +02:00
import java.util.concurrent.ConcurrentHashMap ;
2006-09-30 00:27:20 +02:00
2008-07-05 02:35:20 +02:00
import de.anomic.http.JakartaCommonsHttpClient ;
import de.anomic.http.JakartaCommonsHttpResponse ;
2008-08-25 20:11:47 +02:00
import de.anomic.http.httpRequestHeader ;
2008-07-10 02:47:37 +02:00
import de.anomic.kelondro.kelondroBLOB ;
import de.anomic.kelondro.kelondroBLOBHeap ;
2005-09-14 12:10:49 +02:00
import de.anomic.kelondro.kelondroException ;
2008-07-11 09:34:48 +02:00
import de.anomic.kelondro.kelondroMap ;
2007-05-09 19:59:36 +02:00
import de.anomic.kelondro.kelondroNaturalOrder ;
2009-01-30 15:48:11 +01:00
import de.anomic.kelondro.kelondroDate ;
import de.anomic.kelondro.kelondroByteBuffer ;
2008-07-05 02:35:20 +02:00
import de.anomic.server.serverFileUtils ;
import de.anomic.server.logging.serverLog ;
import de.anomic.yacy.yacyURL ;
2005-09-07 13:17:21 +02:00
2008-05-06 02:32:41 +02:00
public class RobotsTxt {
2006-03-09 15:03:54 +01:00
public static final String ROBOTS_DB_PATH_SEPARATOR = " ; " ;
2008-09-03 02:30:21 +02:00
private static final serverLog log = new serverLog ( " ROBOTS " ) ;
2006-03-09 15:03:54 +01:00
2008-07-11 09:34:48 +02:00
kelondroMap robotsTable ;
2005-10-09 06:43:07 +02:00
private final File robotsTableFile ;
2008-08-02 14:12:04 +02:00
private final ConcurrentHashMap < String , Long > syncObjects ;
2008-07-11 11:12:54 +02:00
//private static final HashSet<String> loadedRobots = new HashSet<String>(); // only for debugging
2005-09-07 13:17:21 +02:00
2008-08-02 14:12:04 +02:00
public RobotsTxt ( final File robotsTableFile ) {
2005-09-07 13:17:21 +02:00
this . robotsTableFile = robotsTableFile ;
2006-08-24 04:19:25 +02:00
robotsTableFile . getParentFile ( ) . mkdirs ( ) ;
2008-07-10 02:47:37 +02:00
kelondroBLOB blob = null ;
2008-12-18 01:18:37 +01:00
try {
blob = new kelondroBLOBHeap ( robotsTableFile , 64 , kelondroNaturalOrder . naturalOrder , 1024 * 1024 ) ;
} catch ( final IOException e ) {
e . printStackTrace ( ) ;
2008-07-10 02:47:37 +02:00
}
2008-07-11 09:34:48 +02:00
robotsTable = new kelondroMap ( blob , 100 ) ;
2008-07-11 14:03:18 +02:00
syncObjects = new ConcurrentHashMap < String , Long > ( ) ;
2006-05-09 12:03:12 +02:00
}
2005-09-07 13:17:21 +02:00
private void resetDatabase ( ) {
// deletes the robots.txt database and creates a new one
2007-03-09 09:48:47 +01:00
if ( robotsTable ! = null ) robotsTable . close ( ) ;
2005-09-07 13:17:21 +02:00
if ( ! ( robotsTableFile . delete ( ) ) ) throw new RuntimeException ( " cannot delete robots.txt database " ) ;
2005-12-13 00:59:58 +01:00
robotsTableFile . getParentFile ( ) . mkdirs ( ) ;
2008-12-18 01:18:37 +01:00
kelondroBLOB blob = null ;
try {
blob = new kelondroBLOBHeap ( robotsTableFile , 64 , kelondroNaturalOrder . naturalOrder , 1024 * 1024 ) ;
} catch ( final IOException e ) {
e . printStackTrace ( ) ;
}
robotsTable = new kelondroMap ( blob , 100 ) ;
syncObjects . clear ( ) ;
2005-09-07 13:17:21 +02:00
}
2008-06-04 23:34:57 +02:00
public void clear ( ) throws IOException {
this . robotsTable . clear ( ) ;
}
2005-09-07 13:17:21 +02:00
public void close ( ) {
2007-05-18 15:00:42 +02:00
this . robotsTable . close ( ) ;
2005-09-07 13:17:21 +02:00
}
public int size ( ) {
2007-05-18 15:00:42 +02:00
return this . robotsTable . size ( ) ;
2005-12-05 15:24:13 +01:00
}
2005-09-07 13:17:21 +02:00
2008-08-02 14:12:04 +02:00
private Entry getEntry ( final String urlHostPort , final boolean fetchOnlineIfNotAvailableOrNotFresh ) {
2008-07-11 11:12:54 +02:00
// this method will always return a non-null value
Entry robotsTxt4Host = null ;
2005-09-07 13:17:21 +02:00
try {
2008-10-10 10:39:11 +02:00
final Map < String , String > record = this . robotsTable . get ( urlHostPort ) ;
2008-07-11 11:12:54 +02:00
if ( record ! = null ) robotsTxt4Host = new Entry ( urlHostPort , record ) ;
2008-08-02 14:12:04 +02:00
} catch ( final kelondroException e ) {
2007-01-30 00:51:10 +01:00
resetDatabase ( ) ;
2008-08-02 14:12:04 +02:00
} catch ( final IOException e ) {
2008-07-11 09:15:46 +02:00
resetDatabase ( ) ;
2005-09-07 13:17:21 +02:00
}
2008-07-11 11:12:54 +02:00
if ( fetchOnlineIfNotAvailableOrNotFresh & & (
robotsTxt4Host = = null | |
robotsTxt4Host . getLoadedDate ( ) = = null | |
System . currentTimeMillis ( ) - robotsTxt4Host . getLoadedDate ( ) . getTime ( ) > 7 * 24 * 60 * 60 * 1000
2008-07-11 14:03:18 +02:00
) ) {
2008-07-11 11:12:54 +02:00
2008-07-11 14:03:18 +02:00
// make or get a synchronization object
Long syncObj = this . syncObjects . get ( urlHostPort ) ;
if ( syncObj = = null ) {
syncObj = new Long ( System . currentTimeMillis ( ) ) ;
this . syncObjects . put ( urlHostPort , syncObj ) ;
2008-07-11 11:12:54 +02:00
}
2008-07-11 14:03:18 +02:00
// we can now synchronize for each host separately
synchronized ( syncObj ) {
// if we have not found any data or the data is older than 7 days, we need to load it from the remote server
// check the robots table again for all threads that come here because they waited for another one
// to complete a download
2008-07-11 11:12:54 +02:00
try {
2008-10-10 10:39:11 +02:00
final Map < String , String > record = this . robotsTable . get ( urlHostPort ) ;
2008-07-11 14:03:18 +02:00
if ( record ! = null ) robotsTxt4Host = new Entry ( urlHostPort , record ) ;
2008-08-02 14:12:04 +02:00
} catch ( final kelondroException e ) {
2008-07-11 14:03:18 +02:00
resetDatabase ( ) ;
2008-08-02 14:12:04 +02:00
} catch ( final IOException e ) {
2008-07-11 14:03:18 +02:00
resetDatabase ( ) ;
2008-07-11 11:12:54 +02:00
}
2008-07-11 14:03:18 +02:00
if ( robotsTxt4Host ! = null & &
robotsTxt4Host . getLoadedDate ( ) ! = null & &
System . currentTimeMillis ( ) - robotsTxt4Host . getLoadedDate ( ) . getTime ( ) < = 7 * 24 * 60 * 60 * 1000 ) {
return robotsTxt4Host ;
}
// generating the proper url to download the robots txt
yacyURL robotsURL = null ;
try {
robotsURL = new yacyURL ( " http:// " + urlHostPort + " /robots.txt " , null ) ;
2008-08-02 14:12:04 +02:00
} catch ( final MalformedURLException e ) {
2008-09-03 02:30:21 +02:00
log . logSevere ( " Unable to generate robots.txt URL for host:port ' " + urlHostPort + " '. " ) ;
2008-07-11 14:03:18 +02:00
robotsURL = null ;
2008-07-11 11:12:54 +02:00
}
2008-07-11 14:03:18 +02:00
Object [ ] result = null ;
if ( robotsURL ! = null ) {
2008-09-03 02:30:21 +02:00
if ( log . isFine ( ) ) log . logFine ( " Trying to download the robots.txt file from URL ' " + robotsURL + " '. " ) ;
2008-07-11 14:03:18 +02:00
try {
result = downloadRobotsTxt ( robotsURL , 5 , robotsTxt4Host ) ;
2008-08-02 14:12:04 +02:00
} catch ( final Exception e ) {
2008-07-11 14:03:18 +02:00
result = null ;
}
2008-07-11 11:12:54 +02:00
}
2008-07-11 14:03:18 +02:00
/ *
assert ! loadedRobots . contains ( robotsURL . toNormalform ( false , false ) ) :
" robots-url= " + robotsURL . toString ( ) +
" , robots= " + ( ( result = = null | | result [ DOWNLOAD_ROBOTS_TXT ] = = null ) ? " NULL " : new String ( ( byte [ ] ) result [ DOWNLOAD_ROBOTS_TXT ] ) ) +
" , robotsTxt4Host= " + ( ( robotsTxt4Host = = null ) ? " NULL " : robotsTxt4Host . getLoadedDate ( ) . toString ( ) ) ;
loadedRobots . add ( robotsURL . toNormalform ( false , false ) ) ;
* /
2008-07-11 11:12:54 +02:00
2008-07-11 14:03:18 +02:00
if ( result = = null ) {
// no robots.txt available, make an entry to prevent that the robots loading is done twice
if ( robotsTxt4Host = = null ) {
// generate artificial entry
robotsTxt4Host = new Entry (
urlHostPort ,
new ArrayList < String > ( ) ,
2008-07-24 13:54:37 +02:00
new ArrayList < String > ( ) ,
2008-07-11 14:03:18 +02:00
new Date ( ) ,
new Date ( ) ,
null ,
null ,
2008-08-06 21:43:12 +02:00
Integer . valueOf ( 0 ) ) ;
2008-07-11 14:03:18 +02:00
} else {
robotsTxt4Host . setLoadedDate ( new Date ( ) ) ;
}
// store the data into the robots DB
addEntry ( robotsTxt4Host ) ;
} else {
2008-08-02 14:12:04 +02:00
final robotsParser parserResult = new robotsParser ( ( byte [ ] ) result [ DOWNLOAD_ROBOTS_TXT ] ) ;
2008-07-24 13:54:37 +02:00
ArrayList < String > denyPath = parserResult . denyList ( ) ;
2008-07-11 14:03:18 +02:00
if ( ( ( Boolean ) result [ DOWNLOAD_ACCESS_RESTRICTED ] ) . booleanValue ( ) ) {
denyPath = new ArrayList < String > ( ) ;
denyPath . add ( " / " ) ;
}
// store the data into the robots DB
robotsTxt4Host = addEntry (
urlHostPort ,
2008-07-24 13:54:37 +02:00
parserResult . allowList ( ) ,
2008-07-11 14:03:18 +02:00
denyPath ,
new Date ( ) ,
( Date ) result [ DOWNLOAD_MODDATE ] ,
( String ) result [ DOWNLOAD_ETAG ] ,
2008-07-24 13:54:37 +02:00
parserResult . sitemap ( ) ,
2008-09-03 01:49:48 +02:00
parserResult . crawlDelayMillis ( ) ) ;
2008-07-11 14:03:18 +02:00
}
2008-07-11 11:12:54 +02:00
}
}
return robotsTxt4Host ;
2008-07-05 02:35:20 +02:00
}
2008-09-03 01:49:48 +02:00
public long crawlDelayMillis ( final yacyURL theURL ) {
2008-08-02 14:12:04 +02:00
final String urlHostPort = getHostPort ( theURL ) ;
final RobotsTxt . Entry robotsEntry = getEntry ( urlHostPort , true ) ;
2008-09-03 01:49:48 +02:00
return robotsEntry . getCrawlDelayMillis ( ) ;
2008-07-05 02:35:20 +02:00
}
2005-09-07 13:17:21 +02:00
2008-07-05 02:35:20 +02:00
private Entry addEntry (
2008-08-02 14:12:04 +02:00
final String hostName ,
final ArrayList < String > allowPathList ,
final ArrayList < String > denyPathList ,
final Date loadedDate ,
final Date modDate ,
final String eTag ,
final String sitemap ,
2008-09-03 01:49:48 +02:00
final long crawlDelayMillis
2007-05-18 15:00:42 +02:00
) {
2008-08-02 14:12:04 +02:00
final Entry entry = new Entry (
2008-07-24 13:54:37 +02:00
hostName , allowPathList , denyPathList , loadedDate , modDate ,
2008-09-03 01:49:48 +02:00
eTag , sitemap , crawlDelayMillis ) ;
2005-09-07 13:17:21 +02:00
addEntry ( entry ) ;
return entry ;
}
2008-08-02 14:12:04 +02:00
private String addEntry ( final Entry entry ) {
2005-09-07 13:17:21 +02:00
// writes a new page and returns key
try {
2008-07-11 00:08:16 +02:00
this . robotsTable . put ( entry . hostName , entry . mem ) ;
2005-09-07 13:17:21 +02:00
return entry . hostName ;
2008-08-02 14:12:04 +02:00
} catch ( final IOException e ) {
2005-09-07 13:17:21 +02:00
return null ;
}
}
2008-08-06 21:43:12 +02:00
public static class Entry {
2008-07-24 13:54:37 +02:00
public static final String ALLOW_PATH_LIST = " allow " ;
2005-09-07 13:17:21 +02:00
public static final String DISALLOW_PATH_LIST = " disallow " ;
2008-07-24 13:54:37 +02:00
public static final String LOADED_DATE = " date " ;
public static final String MOD_DATE = " modDate " ;
public static final String ETAG = " etag " ;
public static final String SITEMAP = " sitemap " ;
public static final String CRAWL_DELAY = " crawlDelay " ;
2008-09-03 01:49:48 +02:00
public static final String CRAWL_DELAY_MILLIS = " crawlDelayMillis " ;
2005-09-07 13:17:21 +02:00
2008-07-24 13:54:37 +02:00
// this is a simple record structure that holds all properties of a single crawl start
2008-10-10 10:39:11 +02:00
Map < String , String > mem ;
2008-07-24 13:54:37 +02:00
private LinkedList < String > allowPathList , denyPathList ;
2007-05-18 15:00:42 +02:00
String hostName ;
2005-09-07 13:17:21 +02:00
2008-10-10 10:39:11 +02:00
public Entry ( final String hostName , final Map < String , String > mem ) {
2005-09-26 10:05:59 +02:00
this . hostName = hostName . toLowerCase ( ) ;
2005-09-07 13:17:21 +02:00
this . mem = mem ;
if ( this . mem . containsKey ( DISALLOW_PATH_LIST ) ) {
2008-07-24 13:54:37 +02:00
this . denyPathList = new LinkedList < String > ( ) ;
2008-08-02 14:12:04 +02:00
final String csPl = this . mem . get ( DISALLOW_PATH_LIST ) ;
2005-09-07 13:17:21 +02:00
if ( csPl . length ( ) > 0 ) {
2008-08-02 14:12:04 +02:00
final String [ ] pathArray = csPl . split ( ROBOTS_DB_PATH_SEPARATOR ) ;
2005-09-07 13:17:21 +02:00
if ( ( pathArray ! = null ) & & ( pathArray . length > 0 ) ) {
2008-07-24 13:54:37 +02:00
this . denyPathList . addAll ( Arrays . asList ( pathArray ) ) ;
}
}
} else {
this . denyPathList = new LinkedList < String > ( ) ;
}
if ( this . mem . containsKey ( ALLOW_PATH_LIST ) ) {
this . allowPathList = new LinkedList < String > ( ) ;
2008-08-02 14:12:04 +02:00
final String csPl = this . mem . get ( ALLOW_PATH_LIST ) ;
2008-07-24 13:54:37 +02:00
if ( csPl . length ( ) > 0 ) {
2008-08-02 14:12:04 +02:00
final String [ ] pathArray = csPl . split ( ROBOTS_DB_PATH_SEPARATOR ) ;
2008-07-24 13:54:37 +02:00
if ( ( pathArray ! = null ) & & ( pathArray . length > 0 ) ) {
this . allowPathList . addAll ( Arrays . asList ( pathArray ) ) ;
2005-09-07 13:17:21 +02:00
}
}
} else {
2008-07-24 13:54:37 +02:00
this . allowPathList = new LinkedList < String > ( ) ;
2005-09-07 13:17:21 +02:00
}
}
2005-10-05 12:45:33 +02:00
public Entry (
2008-08-02 14:12:04 +02:00
final String hostName ,
final ArrayList < String > allowPathList ,
final ArrayList < String > disallowPathList ,
final Date loadedDate ,
final Date modDate ,
final String eTag ,
final String sitemap ,
2008-09-03 01:49:48 +02:00
final long crawlDelayMillis
2007-05-18 15:00:42 +02:00
) {
if ( ( hostName = = null ) | | ( hostName . length ( ) = = 0 ) ) throw new IllegalArgumentException ( " The hostname is missing " ) ;
2005-09-07 13:17:21 +02:00
this . hostName = hostName . trim ( ) . toLowerCase ( ) ;
2008-07-24 13:54:37 +02:00
this . allowPathList = new LinkedList < String > ( ) ;
this . denyPathList = new LinkedList < String > ( ) ;
2005-09-07 13:17:21 +02:00
2008-01-24 23:49:00 +01:00
this . mem = new HashMap < String , String > ( 5 ) ;
2005-09-07 13:17:21 +02:00
if ( loadedDate ! = null ) this . mem . put ( LOADED_DATE , Long . toString ( loadedDate . getTime ( ) ) ) ;
2005-10-05 12:45:33 +02:00
if ( modDate ! = null ) this . mem . put ( MOD_DATE , Long . toString ( modDate . getTime ( ) ) ) ;
if ( eTag ! = null ) this . mem . put ( ETAG , eTag ) ;
2007-04-26 17:42:38 +02:00
if ( sitemap ! = null ) this . mem . put ( SITEMAP , sitemap ) ;
2008-09-03 01:49:48 +02:00
if ( crawlDelayMillis > 0 ) this . mem . put ( CRAWL_DELAY_MILLIS , Long . toString ( crawlDelayMillis ) ) ;
2008-07-24 13:54:37 +02:00
if ( ( allowPathList ! = null ) & & ( allowPathList . size ( ) > 0 ) ) {
this . allowPathList . addAll ( allowPathList ) ;
2008-12-04 13:54:16 +01:00
final StringBuilder pathListStr = new StringBuilder ( ) ;
2008-07-24 13:54:37 +02:00
for ( int i = 0 ; i < allowPathList . size ( ) ; i + + ) {
pathListStr . append ( allowPathList . get ( i ) )
. append ( ROBOTS_DB_PATH_SEPARATOR ) ;
}
this . mem . put ( ALLOW_PATH_LIST , pathListStr . substring ( 0 , pathListStr . length ( ) - 1 ) ) ;
}
2005-09-07 13:17:21 +02:00
if ( ( disallowPathList ! = null ) & & ( disallowPathList . size ( ) > 0 ) ) {
2008-07-24 13:54:37 +02:00
this . denyPathList . addAll ( disallowPathList ) ;
2005-09-07 13:17:21 +02:00
2008-12-04 13:54:16 +01:00
final StringBuilder pathListStr = new StringBuilder ( ) ;
2005-09-07 13:17:21 +02:00
for ( int i = 0 ; i < disallowPathList . size ( ) ; i + + ) {
pathListStr . append ( disallowPathList . get ( i ) )
2006-03-09 15:03:54 +01:00
. append ( ROBOTS_DB_PATH_SEPARATOR ) ;
2005-09-07 13:17:21 +02:00
}
this . mem . put ( DISALLOW_PATH_LIST , pathListStr . substring ( 0 , pathListStr . length ( ) - 1 ) ) ;
}
}
public String toString ( ) {
2008-12-04 13:54:16 +01:00
final StringBuilder str = new StringBuilder ( ) ;
2005-09-07 13:17:21 +02:00
str . append ( ( this . hostName = = null ) ? " null " : this . hostName )
. append ( " : " ) ;
if ( this . mem ! = null ) {
str . append ( this . mem . toString ( ) ) ;
}
return str . toString ( ) ;
}
2007-04-26 17:42:38 +02:00
public String getSitemap ( ) {
2007-05-06 11:52:04 +02:00
return this . mem . containsKey ( SITEMAP ) ? ( String ) this . mem . get ( SITEMAP ) : null ;
2007-04-26 17:42:38 +02:00
}
2005-09-07 13:17:21 +02:00
public Date getLoadedDate ( ) {
if ( this . mem . containsKey ( LOADED_DATE ) ) {
2008-06-06 18:01:27 +02:00
return new Date ( Long . valueOf ( this . mem . get ( LOADED_DATE ) ) . longValue ( ) ) ;
2005-09-07 13:17:21 +02:00
}
return null ;
}
2005-10-18 09:45:27 +02:00
2008-08-02 14:12:04 +02:00
public void setLoadedDate ( final Date newLoadedDate ) {
2005-10-18 09:45:27 +02:00
if ( newLoadedDate ! = null ) {
this . mem . put ( LOADED_DATE , Long . toString ( newLoadedDate . getTime ( ) ) ) ;
}
}
2005-09-07 13:17:21 +02:00
2005-10-05 12:45:33 +02:00
public Date getModDate ( ) {
if ( this . mem . containsKey ( MOD_DATE ) ) {
2008-06-06 18:01:27 +02:00
return new Date ( Long . valueOf ( this . mem . get ( MOD_DATE ) ) . longValue ( ) ) ;
2005-10-05 12:45:33 +02:00
}
return null ;
}
public String getETag ( ) {
if ( this . mem . containsKey ( ETAG ) ) {
2008-06-06 18:01:27 +02:00
return this . mem . get ( ETAG ) ;
2005-10-05 12:45:33 +02:00
}
return null ;
}
2008-09-03 01:49:48 +02:00
public long getCrawlDelayMillis ( ) {
if ( this . mem . containsKey ( CRAWL_DELAY_MILLIS ) ) try {
return Long . parseLong ( this . mem . get ( CRAWL_DELAY_MILLIS ) ) ;
} catch ( final NumberFormatException e ) {
return 0 ;
}
2008-07-11 11:12:54 +02:00
if ( this . mem . containsKey ( CRAWL_DELAY ) ) try {
2008-09-03 01:49:48 +02:00
return 1000 * Integer . parseInt ( this . mem . get ( CRAWL_DELAY ) ) ;
2008-08-02 14:12:04 +02:00
} catch ( final NumberFormatException e ) {
2008-07-11 11:12:54 +02:00
return 0 ;
2007-05-18 15:00:42 +02:00
}
2008-07-11 11:12:54 +02:00
return 0 ;
2007-05-18 15:00:42 +02:00
}
2005-09-07 13:17:21 +02:00
public boolean isDisallowed ( String path ) {
2008-07-24 13:54:37 +02:00
if ( ( this . mem = = null ) | | ( this . denyPathList . size ( ) = = 0 ) ) return false ;
2006-03-09 15:03:54 +01:00
// if the path is null or empty we set it to /
if ( ( path = = null ) | | ( path . length ( ) = = 0 ) ) path = " / " ;
// escaping all occurences of ; because this char is used as special char in the Robots DB
else path = path . replaceAll ( ROBOTS_DB_PATH_SEPARATOR , " %3B " ) ;
2008-08-02 14:12:04 +02:00
final Iterator < String > pathIter = this . denyPathList . iterator ( ) ;
2005-09-07 13:17:21 +02:00
while ( pathIter . hasNext ( ) ) {
2008-08-02 14:12:04 +02:00
final String nextPath = pathIter . next ( ) ;
2006-03-09 15:03:54 +01:00
// disallow rule
if ( path . startsWith ( nextPath ) ) {
return true ;
}
2005-09-07 13:17:21 +02:00
}
return false ;
}
}
2008-07-05 02:35:20 +02:00
// methods that had been in robotsParser.java:
public static final int DOWNLOAD_ACCESS_RESTRICTED = 0 ;
public static final int DOWNLOAD_ROBOTS_TXT = 1 ;
public static final int DOWNLOAD_ETAG = 2 ;
public static final int DOWNLOAD_MODDATE = 3 ;
2008-08-02 14:12:04 +02:00
private static final String getHostPort ( final yacyURL theURL ) {
2008-07-05 02:35:20 +02:00
String urlHostPort = null ;
2008-08-02 14:12:04 +02:00
final int port = getPort ( theURL ) ;
2008-07-05 02:35:20 +02:00
urlHostPort = theURL . getHost ( ) + " : " + port ;
urlHostPort = urlHostPort . toLowerCase ( ) . intern ( ) ;
return urlHostPort ;
}
2008-08-02 14:12:04 +02:00
private static final int getPort ( final yacyURL theURL ) {
2008-07-05 02:35:20 +02:00
int port = theURL . getPort ( ) ;
if ( port = = - 1 ) {
if ( theURL . getProtocol ( ) . equalsIgnoreCase ( " http " ) ) {
port = 80 ;
} else if ( theURL . getProtocol ( ) . equalsIgnoreCase ( " https " ) ) {
port = 443 ;
}
}
return port ;
}
2008-08-02 14:12:04 +02:00
public yacyURL getSitemapURL ( final yacyURL theURL ) {
2008-07-05 02:35:20 +02:00
if ( theURL = = null ) throw new IllegalArgumentException ( ) ;
yacyURL sitemapURL = null ;
// generating the hostname:poart string needed to do a DB lookup
2008-08-02 14:12:04 +02:00
final String urlHostPort = getHostPort ( theURL ) ;
final RobotsTxt . Entry robotsTxt4Host = this . getEntry ( urlHostPort , true ) ;
2008-07-05 02:35:20 +02:00
try {
2008-08-02 14:12:04 +02:00
final String sitemapUrlStr = robotsTxt4Host . getSitemap ( ) ;
2008-07-05 02:35:20 +02:00
if ( sitemapUrlStr ! = null ) sitemapURL = new yacyURL ( sitemapUrlStr , null ) ;
2008-08-02 14:12:04 +02:00
} catch ( final MalformedURLException e ) { /* ignore this */ }
2008-07-05 02:35:20 +02:00
return sitemapURL ;
}
2008-09-03 01:49:48 +02:00
public Long getCrawlDelayMillis ( final yacyURL theURL ) {
2008-07-05 02:35:20 +02:00
if ( theURL = = null ) throw new IllegalArgumentException ( ) ;
2008-09-03 01:49:48 +02:00
Long crawlDelay = null ;
2008-07-05 02:35:20 +02:00
// generating the hostname:poart string needed to do a DB lookup
2008-08-02 14:12:04 +02:00
final String urlHostPort = getHostPort ( theURL ) ;
final RobotsTxt . Entry robotsTxt4Host = getEntry ( urlHostPort , true ) ;
2008-07-05 02:35:20 +02:00
try {
2008-09-03 01:49:48 +02:00
crawlDelay = robotsTxt4Host . getCrawlDelayMillis ( ) ;
2008-08-02 14:12:04 +02:00
} catch ( final NumberFormatException e ) { /* ignore this */ }
2008-07-05 02:35:20 +02:00
2008-09-03 01:49:48 +02:00
return crawlDelay ;
2008-07-05 02:35:20 +02:00
}
2008-08-02 14:12:04 +02:00
public boolean isDisallowed ( final yacyURL nexturl ) {
2008-07-05 02:35:20 +02:00
if ( nexturl = = null ) throw new IllegalArgumentException ( ) ;
2008-07-11 11:12:54 +02:00
// generating the hostname:port string needed to do a DB lookup
2008-08-02 14:12:04 +02:00
final String urlHostPort = getHostPort ( nexturl ) ;
2008-07-10 02:47:37 +02:00
RobotsTxt . Entry robotsTxt4Host = null ;
2008-07-11 11:12:54 +02:00
robotsTxt4Host = getEntry ( urlHostPort , true ) ;
2008-07-10 02:47:37 +02:00
return robotsTxt4Host . isDisallowed ( nexturl . getFile ( ) ) ;
2008-07-05 02:35:20 +02:00
}
2005-09-07 13:17:21 +02:00
2008-08-02 14:12:04 +02:00
private static Object [ ] downloadRobotsTxt ( final yacyURL robotsURL , int redirectionCount , final RobotsTxt . Entry entry ) throws Exception {
2008-07-05 02:35:20 +02:00
if ( redirectionCount < 0 ) return new Object [ ] { Boolean . FALSE , null , null } ;
redirectionCount - - ;
boolean accessCompletelyRestricted = false ;
byte [ ] robotsTxt = null ;
long downloadStart , downloadEnd ;
String eTag = null , oldEtag = null ;
Date lastMod = null ;
downloadStart = System . currentTimeMillis ( ) ;
// if we previously have downloaded this robots.txt then we can set the if-modified-since header
2008-08-25 20:11:47 +02:00
httpRequestHeader reqHeaders = new httpRequestHeader ( ) ;
2008-07-05 02:35:20 +02:00
// add yacybot user agent
2008-08-25 20:11:47 +02:00
reqHeaders . put ( httpRequestHeader . USER_AGENT , HTTPLoader . crawlerUserAgent ) ;
2008-07-05 02:35:20 +02:00
// adding referer
2008-08-25 20:11:47 +02:00
reqHeaders . put ( httpRequestHeader . REFERER , ( yacyURL . newURL ( robotsURL , " / " ) ) . toNormalform ( true , true ) ) ;
2008-07-05 02:35:20 +02:00
if ( entry ! = null ) {
oldEtag = entry . getETag ( ) ;
2008-08-25 20:11:47 +02:00
reqHeaders = new httpRequestHeader ( ) ;
2008-08-02 14:12:04 +02:00
final Date modDate = entry . getModDate ( ) ;
2009-01-30 15:48:11 +01:00
if ( modDate ! = null ) reqHeaders . put ( httpRequestHeader . IF_MODIFIED_SINCE , kelondroDate . formatRFC1123 ( entry . getModDate ( ) ) ) ;
2008-07-05 02:35:20 +02:00
}
// setup http-client
//TODO: adding Traffic statistic for robots download?
2008-08-17 12:16:32 +02:00
final JakartaCommonsHttpClient client = new JakartaCommonsHttpClient ( 10000 , reqHeaders ) ;
2008-07-05 02:35:20 +02:00
JakartaCommonsHttpResponse res = null ;
try {
// sending the get request
res = client . GET ( robotsURL . toString ( ) ) ;
// check for interruption
if ( Thread . currentThread ( ) . isInterrupted ( ) ) throw new InterruptedException ( " Shutdown in progress. " ) ;
// check the response status
if ( res . getStatusLine ( ) . startsWith ( " 2 " ) ) {
if ( ! res . getResponseHeader ( ) . mime ( ) . startsWith ( " text/plain " ) ) {
robotsTxt = null ;
2008-09-03 02:30:21 +02:00
if ( log . isFinest ( ) ) log . logFinest ( " Robots.txt from URL ' " + robotsURL + " ' has wrong mimetype ' " + res . getResponseHeader ( ) . mime ( ) + " '. " ) ;
2008-07-05 02:35:20 +02:00
} else {
// getting some metadata
2008-08-25 20:11:47 +02:00
eTag = res . getResponseHeader ( ) . containsKey ( httpRequestHeader . ETAG ) ? ( res . getResponseHeader ( ) . get ( httpRequestHeader . ETAG ) ) . trim ( ) : null ;
2008-07-05 02:35:20 +02:00
lastMod = res . getResponseHeader ( ) . lastModified ( ) ;
// if the robots.txt file was not changed we break here
if ( ( eTag ! = null ) & & ( oldEtag ! = null ) & & ( eTag . equals ( oldEtag ) ) ) {
2008-09-03 02:30:21 +02:00
if ( log . isFinest ( ) ) log . logFinest ( " Robots.txt from URL ' " + robotsURL + " ' was not modified. Abort downloading of new version. " ) ;
2008-07-05 02:35:20 +02:00
return null ;
}
// downloading the content
2009-01-30 15:48:11 +01:00
final kelondroByteBuffer sbb = new kelondroByteBuffer ( ) ;
2008-07-05 02:35:20 +02:00
try {
serverFileUtils . copyToStream ( new BufferedInputStream ( res . getDataAsStream ( ) ) , new BufferedOutputStream ( sbb ) ) ;
} finally {
res . closeStream ( ) ;
}
robotsTxt = sbb . getBytes ( ) ;
downloadEnd = System . currentTimeMillis ( ) ;
2008-09-03 02:30:21 +02:00
if ( log . isFinest ( ) ) log . logFinest ( " Robots.txt successfully loaded from URL ' " + robotsURL + " ' in " + ( downloadEnd - downloadStart ) + " ms. " ) ;
2008-07-05 02:35:20 +02:00
}
} else if ( res . getStatusCode ( ) = = 304 ) {
return null ;
} else if ( res . getStatusLine ( ) . startsWith ( " 3 " ) ) {
// getting redirection URL
2008-08-25 20:11:47 +02:00
String redirectionUrlString = res . getResponseHeader ( ) . get ( httpRequestHeader . LOCATION ) ;
2008-07-05 02:35:20 +02:00
if ( redirectionUrlString = = null ) {
2008-09-03 02:30:21 +02:00
if ( log . isFinest ( ) ) log . logFinest ( " robots.txt could not be downloaded from URL ' " + robotsURL + " ' because of missing redirecton header. [ " + res . getStatusLine ( ) + " ]. " ) ;
2008-07-05 02:35:20 +02:00
robotsTxt = null ;
} else {
redirectionUrlString = redirectionUrlString . trim ( ) ;
// generating the new URL object
2008-08-02 14:12:04 +02:00
final yacyURL redirectionUrl = yacyURL . newURL ( robotsURL , redirectionUrlString ) ;
2008-07-05 02:35:20 +02:00
// following the redirection
2008-09-03 02:30:21 +02:00
if ( log . isFinest ( ) ) log . logFinest ( " Redirection detected for robots.txt with URL ' " + robotsURL + " '. " +
2008-07-05 02:35:20 +02:00
" \ nRedirecting request to: " + redirectionUrl ) ;
return downloadRobotsTxt ( redirectionUrl , redirectionCount , entry ) ;
}
} else if ( res . getStatusCode ( ) = = 401 | | res . getStatusCode ( ) = = 403 ) {
accessCompletelyRestricted = true ;
2008-09-03 02:30:21 +02:00
if ( log . isFinest ( ) ) log . logFinest ( " Access to Robots.txt not allowed on URL ' " + robotsURL + " '. " ) ;
2008-07-05 02:35:20 +02:00
} else {
2008-09-03 02:30:21 +02:00
if ( log . isFinest ( ) ) log . logFinest ( " robots.txt could not be downloaded from URL ' " + robotsURL + " '. [ " + res . getStatusLine ( ) + " ]. " ) ;
2008-07-05 02:35:20 +02:00
robotsTxt = null ;
}
2008-08-02 14:12:04 +02:00
} catch ( final Exception e ) {
2008-07-05 02:35:20 +02:00
throw e ;
} finally {
if ( res ! = null ) {
// release connection
res . closeStream ( ) ;
}
}
2008-08-06 21:43:12 +02:00
return new Object [ ] { Boolean . valueOf ( accessCompletelyRestricted ) , robotsTxt , eTag , lastMod } ;
2008-07-05 02:35:20 +02:00
}
2005-09-07 13:17:21 +02:00
}