Author: jnioche
Date: Mon Nov 18 12:08:45 2013
New Revision: 1543010
URL: http://svn.apache.org/r1543010
Log:
NUTCH-1668 Remove package org.apache.nutch.indexer.solr
Removed:
nutch/trunk/src/java/org/apache/nutch/indexer/solr/
Modified:
nutch/trunk/CHANGES.txt
nutch/trunk/ivy/ivy.xml
nutch/trunk/src/bin/nutch
nutch/trunk/src/java/org/apache/nutch/indexer/IndexingJob.java
Modified: nutch/trunk/CHANGES.txt
URL:
http://svn.apache.org/viewvc/nutch/trunk/CHANGES.txt?rev=1543010&r1=1543009&r2=1543010&view=diff
==============================================================================
--- nutch/trunk/CHANGES.txt (original)
+++ nutch/trunk/CHANGES.txt Mon Nov 18 12:08:45 2013
@@ -2,6 +2,8 @@ Nutch Change Log
Nutch Development Trunk
+* NUTCH-1668 Remove package org.apache.nutch.indexer.solr (jnioche)
+
* NUTCH-1621 Remove deprecated class o.a.n.crawl.Crawler (Rui Gao via jnioche)
* NUTCH-656 Generic Deduplicator (jnioche, snagel)
Modified: nutch/trunk/ivy/ivy.xml
URL:
http://svn.apache.org/viewvc/nutch/trunk/ivy/ivy.xml?rev=1543010&r1=1543009&r2=1543010&view=diff
==============================================================================
--- nutch/trunk/ivy/ivy.xml (original)
+++ nutch/trunk/ivy/ivy.xml Mon Nov 18 12:08:45 2013
@@ -34,11 +34,8 @@
</publications>
<dependencies>
- <dependency org="org.elasticsearch" name="elasticsearch"
rev="0.90.1"
- conf="*->default"/>
-
- <dependency org="org.apache.solr" name="solr-solrj" rev="3.4.0"
- conf="*->default" />
+ <dependency org="org.elasticsearch" name="elasticsearch"
+ rev="0.90.1" conf="*->default" />
<dependency org="org.slf4j" name="slf4j-api" rev="1.6.1"
conf="*->master" />
@@ -77,7 +74,8 @@
<dependency org="oro" name="oro" rev="2.0.8" />
<dependency org="com.google.guava" name="guava" rev="11.0.2" />
- <dependency org="com.google.code.crawler-commons"
name="crawler-commons" rev="0.2" />
+ <dependency org="com.google.code.crawler-commons"
name="crawler-commons"
+ rev="0.2" />
<!--Configuration: test -->
Modified: nutch/trunk/src/bin/nutch
URL:
http://svn.apache.org/viewvc/nutch/trunk/src/bin/nutch?rev=1543010&r1=1543009&r2=1543010&view=diff
==============================================================================
--- nutch/trunk/src/bin/nutch (original)
+++ nutch/trunk/src/bin/nutch Mon Nov 18 12:08:45 2013
@@ -61,10 +61,11 @@ if [ $# = 0 ]; then
echo " invertlinks create a linkdb from parsed segments"
echo " mergelinkdb merge linkdb-s, with optional filtering"
echo " index run the plugin-based indexer on parsed segments
and linkdb"
- echo " solrindex run the solr indexer on parsed segments and linkdb"
- echo " solrdedup remove duplicates from solr"
- echo " solrclean remove HTTP 301 and 404 documents from solr"
- echo " clean remove HTTP 301 and 404 documents from indexing
backends configured via plugins"
+ echo " dedup deduplicate entries in the crawldb and give them a
special status"
+ echo " solrindex run the solr indexer on parsed segments and linkdb
- DEPRECATED use the index command instead"
+ echo " solrdedup remove duplicates from solr - DEPRECATED use the
dedup command instead"
+ echo " solrclean remove HTTP 301 and 404 documents from solr -
DEPRECATED use the clean command instead"
+ echo " clean remove HTTP 301 and 404 documents and duplicates
from indexing backends configured via plugins"
echo " parsechecker check the parser for a given url"
echo " indexchecker check the indexing filters for a given url"
echo " domainstats calculate domain statistics from crawldb"
@@ -226,7 +227,8 @@ elif [ "$COMMAND" = "solrindex" ] ; then
elif [ "$COMMAND" = "index" ] ; then
CLASS=org.apache.nutch.indexer.IndexingJob
elif [ "$COMMAND" = "solrdedup" ] ; then
- CLASS=org.apache.nutch.indexer.solr.SolrDeleteDuplicates
+ echo "Command $COMMAND is deprecated, please use dedup instead"
+ exit -1
elif [ "$COMMAND" = "dedup" ] ; then
CLASS=org.apache.nutch.crawl.DeduplicationJob
elif [ "$COMMAND" = "solrclean" ] ; then
Modified: nutch/trunk/src/java/org/apache/nutch/indexer/IndexingJob.java
URL:
http://svn.apache.org/viewvc/nutch/trunk/src/java/org/apache/nutch/indexer/IndexingJob.java?rev=1543010&r1=1543009&r2=1543010&view=diff
==============================================================================
--- nutch/trunk/src/java/org/apache/nutch/indexer/IndexingJob.java (original)
+++ nutch/trunk/src/java/org/apache/nutch/indexer/IndexingJob.java Mon Nov 18
12:08:45 2013
@@ -33,7 +33,6 @@ import org.apache.hadoop.mapred.JobConf;
import org.apache.hadoop.util.StringUtils;
import org.apache.hadoop.util.Tool;
import org.apache.hadoop.util.ToolRunner;
-import org.apache.nutch.indexer.solr.SolrConstants;
import org.apache.nutch.util.HadoopFSUtil;
import org.apache.nutch.util.NutchConfiguration;
import org.apache.nutch.util.NutchJob;
@@ -57,14 +56,6 @@ public class IndexingJob extends Configu
super(conf);
}
- public void index(Path crawlDb, Path linkDb, List<Path> segments)
- throws IOException {
- // SOLR SPECIFIC COMMIT PARAM USED BY CRAWL CLASS
- boolean noCommit = !getConf().getBoolean(SolrConstants.COMMIT_INDEX,
- true);
- index(crawlDb, linkDb, segments, noCommit, false, null);
- }
-
public void index(Path crawlDb, Path linkDb, List<Path> segments,
boolean noCommit) throws IOException {
index(crawlDb, linkDb, segments, noCommit, false, null);
