Author: smarthi
Date: Tue Apr 9 16:37:44 2013
New Revision: 1466110
URL: http://svn.apache.org/r1466110
Log:
Mahout-1188: Inconsistency in Lucene versions between codebase and POM
Modified:
mahout/trunk/CHANGELOG
mahout/trunk/core/src/main/java/org/apache/mahout/common/lucene/AnalyzerUtils.java
mahout/trunk/core/src/test/java/org/apache/mahout/vectorizer/encoders/TextValueEncoderTest.java
mahout/trunk/examples/src/main/java/org/apache/mahout/classifier/NewsgroupHelper.java
mahout/trunk/integration/src/main/java/org/apache/mahout/text/MailArchivesClusteringAnalyzer.java
mahout/trunk/integration/src/main/java/org/apache/mahout/text/wikipedia/WikipediaAnalyzer.java
mahout/trunk/integration/src/main/java/org/apache/mahout/utils/regex/AnalyzerTransformer.java
mahout/trunk/integration/src/test/java/org/apache/mahout/clustering/TestClusterDumper.java
mahout/trunk/integration/src/test/java/org/apache/mahout/clustering/dirichlet/TestL1ModelClustering.java
mahout/trunk/integration/src/test/java/org/apache/mahout/utils/nlp/collocations/llr/BloomTokenFilterTest.java
mahout/trunk/integration/src/test/java/org/apache/mahout/utils/vectors/lucene/CachedTermInfoTest.java
mahout/trunk/integration/src/test/java/org/apache/mahout/utils/vectors/lucene/LuceneIterableTest.java
mahout/trunk/pom.xml
Modified: mahout/trunk/CHANGELOG
URL:
http://svn.apache.org/viewvc/mahout/trunk/CHANGELOG?rev=1466110&r1=1466109&r2=1466110&view=diff
==============================================================================
--- mahout/trunk/CHANGELOG (original)
+++ mahout/trunk/CHANGELOG Tue Apr 9 16:37:44 2013
@@ -2,6 +2,8 @@ Mahout Change Log
Release 0.8 - unreleased
+ MAHOUT-1188: Inconsistent reference to Lucene versions in code and POM
(smarthi)
+
MAHOUT-1161: Unable to run CJKAnalyzer for conversion of a sequence file to
sparse vector due to instantiation
exception (ssc)
Modified:
mahout/trunk/core/src/main/java/org/apache/mahout/common/lucene/AnalyzerUtils.java
URL:
http://svn.apache.org/viewvc/mahout/trunk/core/src/main/java/org/apache/mahout/common/lucene/AnalyzerUtils.java?rev=1466110&r1=1466109&r2=1466110&view=diff
==============================================================================
---
mahout/trunk/core/src/main/java/org/apache/mahout/common/lucene/AnalyzerUtils.java
(original)
+++
mahout/trunk/core/src/main/java/org/apache/mahout/common/lucene/AnalyzerUtils.java
Tue Apr 9 16:37:44 2013
@@ -32,7 +32,7 @@ public final class AnalyzerUtils {
* @throws ClassNotFoundException - {@link ClassNotFoundException}
*/
public static Analyzer createAnalyzer(String analyzerClassName) throws
ClassNotFoundException {
- return createAnalyzer(analyzerClassName, Version.LUCENE_41);
+ return createAnalyzer(analyzerClassName, Version.LUCENE_42);
}
public static Analyzer createAnalyzer(String analyzerClassName, Version
version) throws ClassNotFoundException {
@@ -47,7 +47,7 @@ public final class AnalyzerUtils {
* @return {@link Analyzer}
*/
public static Analyzer createAnalyzer(Class<? extends Analyzer>
analyzerClass) {
- return createAnalyzer(analyzerClass, Version.LUCENE_41);
+ return createAnalyzer(analyzerClass, Version.LUCENE_42);
}
public static Analyzer createAnalyzer(Class<? extends Analyzer>
analyzerClass, Version version) {
Modified:
mahout/trunk/core/src/test/java/org/apache/mahout/vectorizer/encoders/TextValueEncoderTest.java
URL:
http://svn.apache.org/viewvc/mahout/trunk/core/src/test/java/org/apache/mahout/vectorizer/encoders/TextValueEncoderTest.java?rev=1466110&r1=1466109&r2=1466110&view=diff
==============================================================================
---
mahout/trunk/core/src/test/java/org/apache/mahout/vectorizer/encoders/TextValueEncoderTest.java
(original)
+++
mahout/trunk/core/src/test/java/org/apache/mahout/vectorizer/encoders/TextValueEncoderTest.java
Tue Apr 9 16:37:44 2013
@@ -70,7 +70,7 @@ public final class TextValueEncoderTest
@Test
public void testLuceneEncoding() throws Exception {
LuceneTextValueEncoder enc = new LuceneTextValueEncoder("text");
- enc.setAnalyzer(new WhitespaceAnalyzer(Version.LUCENE_41));
+ enc.setAnalyzer(new WhitespaceAnalyzer(Version.LUCENE_42));
Vector v1 = new DenseVector(200);
enc.addToVector("test1 and more", v1);
enc.flush(1, v1);
Modified:
mahout/trunk/examples/src/main/java/org/apache/mahout/classifier/NewsgroupHelper.java
URL:
http://svn.apache.org/viewvc/mahout/trunk/examples/src/main/java/org/apache/mahout/classifier/NewsgroupHelper.java?rev=1466110&r1=1466109&r2=1466110&view=diff
==============================================================================
---
mahout/trunk/examples/src/main/java/org/apache/mahout/classifier/NewsgroupHelper.java
(original)
+++
mahout/trunk/examples/src/main/java/org/apache/mahout/classifier/NewsgroupHelper.java
Tue Apr 9 16:37:44 2013
@@ -60,7 +60,7 @@ public final class NewsgroupHelper {
private static final long WEEK = 7 * 24 * 3600;
private final Random rand = RandomUtils.getRandom();
- private final Analyzer analyzer = new StandardAnalyzer(Version.LUCENE_41);
+ private final Analyzer analyzer = new StandardAnalyzer(Version.LUCENE_42);
private final FeatureVectorEncoder encoder = new
StaticWordValueEncoder("body");
private final FeatureVectorEncoder bias = new
ConstantValueEncoder("Intercept");
Modified:
mahout/trunk/integration/src/main/java/org/apache/mahout/text/MailArchivesClusteringAnalyzer.java
URL:
http://svn.apache.org/viewvc/mahout/trunk/integration/src/main/java/org/apache/mahout/text/MailArchivesClusteringAnalyzer.java?rev=1466110&r1=1466109&r2=1466110&view=diff
==============================================================================
---
mahout/trunk/integration/src/main/java/org/apache/mahout/text/MailArchivesClusteringAnalyzer.java
(original)
+++
mahout/trunk/integration/src/main/java/org/apache/mahout/text/MailArchivesClusteringAnalyzer.java
Tue Apr 9 16:37:44 2013
@@ -41,7 +41,7 @@ import org.apache.lucene.util.Version;
* stop words, excluding non-alpha-numeric tokens, and porter stemming.
*/
public final class MailArchivesClusteringAnalyzer extends StopwordAnalyzerBase
{
- private static final Version LUCENE_VERSION = Version.LUCENE_41;
+ private static final Version LUCENE_VERSION = Version.LUCENE_42;
// extended set of stop words composed of common mail terms like "hi",
// HTML tags, and Java keywords asmany of the messages in the archives
Modified:
mahout/trunk/integration/src/main/java/org/apache/mahout/text/wikipedia/WikipediaAnalyzer.java
URL:
http://svn.apache.org/viewvc/mahout/trunk/integration/src/main/java/org/apache/mahout/text/wikipedia/WikipediaAnalyzer.java?rev=1466110&r1=1466109&r2=1466110&view=diff
==============================================================================
---
mahout/trunk/integration/src/main/java/org/apache/mahout/text/wikipedia/WikipediaAnalyzer.java
(original)
+++
mahout/trunk/integration/src/main/java/org/apache/mahout/text/wikipedia/WikipediaAnalyzer.java
Tue Apr 9 16:37:44 2013
@@ -33,19 +33,19 @@ import org.apache.lucene.util.Version;
public class WikipediaAnalyzer extends StopwordAnalyzerBase {
public WikipediaAnalyzer() {
- super(Version.LUCENE_41, StopAnalyzer.ENGLISH_STOP_WORDS_SET);
+ super(Version.LUCENE_42, StopAnalyzer.ENGLISH_STOP_WORDS_SET);
}
public WikipediaAnalyzer(CharArraySet stopSet) {
- super(Version.LUCENE_41, stopSet);
+ super(Version.LUCENE_42, stopSet);
}
@Override
protected TokenStreamComponents createComponents(String fieldName, Reader
reader) {
Tokenizer tokenizer = new WikipediaTokenizer(reader);
- TokenStream result = new StandardFilter(Version.LUCENE_41, tokenizer);
- result = new LowerCaseFilter(Version.LUCENE_41, result);
- result = new StopFilter(Version.LUCENE_41, result, getStopwordSet());
+ TokenStream result = new StandardFilter(Version.LUCENE_42, tokenizer);
+ result = new LowerCaseFilter(Version.LUCENE_42, result);
+ result = new StopFilter(Version.LUCENE_42, result, getStopwordSet());
return new TokenStreamComponents(tokenizer, result);
}
}
Modified:
mahout/trunk/integration/src/main/java/org/apache/mahout/utils/regex/AnalyzerTransformer.java
URL:
http://svn.apache.org/viewvc/mahout/trunk/integration/src/main/java/org/apache/mahout/utils/regex/AnalyzerTransformer.java?rev=1466110&r1=1466109&r2=1466110&view=diff
==============================================================================
---
mahout/trunk/integration/src/main/java/org/apache/mahout/utils/regex/AnalyzerTransformer.java
(original)
+++
mahout/trunk/integration/src/main/java/org/apache/mahout/utils/regex/AnalyzerTransformer.java
Tue Apr 9 16:37:44 2013
@@ -34,7 +34,7 @@ public class AnalyzerTransformer impleme
private String fieldName = "text";
public AnalyzerTransformer() {
- this(new StandardAnalyzer(Version.LUCENE_41), "text");
+ this(new StandardAnalyzer(Version.LUCENE_42), "text");
}
public AnalyzerTransformer(Analyzer analyzer) {
Modified:
mahout/trunk/integration/src/test/java/org/apache/mahout/clustering/TestClusterDumper.java
URL:
http://svn.apache.org/viewvc/mahout/trunk/integration/src/test/java/org/apache/mahout/clustering/TestClusterDumper.java?rev=1466110&r1=1466109&r2=1466110&view=diff
==============================================================================
---
mahout/trunk/integration/src/test/java/org/apache/mahout/clustering/TestClusterDumper.java
(original)
+++
mahout/trunk/integration/src/test/java/org/apache/mahout/clustering/TestClusterDumper.java
Tue Apr 9 16:37:44 2013
@@ -105,8 +105,8 @@ public final class TestClusterDumper ext
RAMDirectory directory = new RAMDirectory();
IndexWriter writer = new IndexWriter(directory,
- new IndexWriterConfig(Version.LUCENE_41,new StandardAnalyzer(
- Version.LUCENE_41)));
+ new IndexWriterConfig(Version.LUCENE_42,new StandardAnalyzer(
+ Version.LUCENE_42)));
try {
for (int i = 0; i < docs2.length; i++) {
Modified:
mahout/trunk/integration/src/test/java/org/apache/mahout/clustering/dirichlet/TestL1ModelClustering.java
URL:
http://svn.apache.org/viewvc/mahout/trunk/integration/src/test/java/org/apache/mahout/clustering/dirichlet/TestL1ModelClustering.java?rev=1466110&r1=1466109&r2=1466110&view=diff
==============================================================================
---
mahout/trunk/integration/src/test/java/org/apache/mahout/clustering/dirichlet/TestL1ModelClustering.java
(original)
+++
mahout/trunk/integration/src/test/java/org/apache/mahout/clustering/dirichlet/TestL1ModelClustering.java
Tue Apr 9 16:37:44 2013
@@ -135,7 +135,7 @@ public final class TestL1ModelClustering
System.out.println();
sampleData = Lists.newArrayList();
RAMDirectory directory = new RAMDirectory();
- IndexWriter writer = new IndexWriter( directory, new
IndexWriterConfig(Version.LUCENE_41,new StandardAnalyzer(Version.LUCENE_41)));
+ IndexWriter writer = new IndexWriter( directory, new
IndexWriterConfig(Version.LUCENE_42,new StandardAnalyzer(Version.LUCENE_42)));
FieldType customType = new FieldType(TextField.TYPE_NOT_STORED);
customType.setStoreTermVectors(true);
Modified:
mahout/trunk/integration/src/test/java/org/apache/mahout/utils/nlp/collocations/llr/BloomTokenFilterTest.java
URL:
http://svn.apache.org/viewvc/mahout/trunk/integration/src/test/java/org/apache/mahout/utils/nlp/collocations/llr/BloomTokenFilterTest.java?rev=1466110&r1=1466109&r2=1466110&view=diff
==============================================================================
---
mahout/trunk/integration/src/test/java/org/apache/mahout/utils/nlp/collocations/llr/BloomTokenFilterTest.java
(original)
+++
mahout/trunk/integration/src/test/java/org/apache/mahout/utils/nlp/collocations/llr/BloomTokenFilterTest.java
Tue Apr 9 16:37:44 2013
@@ -79,7 +79,7 @@ public final class BloomTokenFilterTest
@Test
public void testAnalyzer() throws IOException {
Reader reader = new StringReader(input);
- Analyzer analyzer = new WhitespaceAnalyzer(Version.LUCENE_41);
+ Analyzer analyzer = new WhitespaceAnalyzer(Version.LUCENE_42);
TokenStream ts = analyzer.tokenStream(null, reader);
ts.reset();
validateTokens(allTokens, ts);
@@ -91,7 +91,7 @@ public final class BloomTokenFilterTest
@Test
public void testNonKeepdAnalyzer() throws IOException {
Reader reader = new StringReader(input);
- Analyzer analyzer = new WhitespaceAnalyzer(Version.LUCENE_41);
+ Analyzer analyzer = new WhitespaceAnalyzer(Version.LUCENE_42);
TokenStream ts = analyzer.tokenStream(null, reader);
ts.reset();
TokenStream f = new BloomTokenFilter(getFilter(filterTokens), false /*
toss matching tokens */, ts);
@@ -104,7 +104,7 @@ public final class BloomTokenFilterTest
@Test
public void testKeepAnalyzer() throws IOException {
Reader reader = new StringReader(input);
- Analyzer analyzer = new WhitespaceAnalyzer(Version.LUCENE_41);
+ Analyzer analyzer = new WhitespaceAnalyzer(Version.LUCENE_42);
TokenStream ts = analyzer.tokenStream(null, reader);
ts.reset();
TokenStream f = new BloomTokenFilter(getFilter(filterTokens), true /* keep
matching tokens */, ts);
@@ -117,7 +117,7 @@ public final class BloomTokenFilterTest
@Test
public void testShingleFilteredAnalyzer() throws IOException {
Reader reader = new StringReader(input);
- Analyzer analyzer = new WhitespaceAnalyzer(Version.LUCENE_41);
+ Analyzer analyzer = new WhitespaceAnalyzer(Version.LUCENE_42);
TokenStream ts = analyzer.tokenStream(null, reader);
ts.reset();
ShingleFilter sf = new ShingleFilter(ts, 3);
Modified:
mahout/trunk/integration/src/test/java/org/apache/mahout/utils/vectors/lucene/CachedTermInfoTest.java
URL:
http://svn.apache.org/viewvc/mahout/trunk/integration/src/test/java/org/apache/mahout/utils/vectors/lucene/CachedTermInfoTest.java?rev=1466110&r1=1466109&r2=1466110&view=diff
==============================================================================
---
mahout/trunk/integration/src/test/java/org/apache/mahout/utils/vectors/lucene/CachedTermInfoTest.java
(original)
+++
mahout/trunk/integration/src/test/java/org/apache/mahout/utils/vectors/lucene/CachedTermInfoTest.java
Tue Apr 9 16:37:44 2013
@@ -76,7 +76,7 @@ public class CachedTermInfoTest extends
RAMDirectory directory,
boolean createNew,
int startingId) throws IOException {
- IndexWriter writer = new IndexWriter(directory, new
IndexWriterConfig(Version.LUCENE_41, new
WhitespaceAnalyzer(Version.LUCENE_41)));
+ IndexWriter writer = new IndexWriter(directory, new
IndexWriterConfig(Version.LUCENE_42, new
WhitespaceAnalyzer(Version.LUCENE_42)));
try {
for (int i = 0; i < DOCS.length; i++) {
Modified:
mahout/trunk/integration/src/test/java/org/apache/mahout/utils/vectors/lucene/LuceneIterableTest.java
URL:
http://svn.apache.org/viewvc/mahout/trunk/integration/src/test/java/org/apache/mahout/utils/vectors/lucene/LuceneIterableTest.java?rev=1466110&r1=1466109&r2=1466110&view=diff
==============================================================================
---
mahout/trunk/integration/src/test/java/org/apache/mahout/utils/vectors/lucene/LuceneIterableTest.java
(original)
+++
mahout/trunk/integration/src/test/java/org/apache/mahout/utils/vectors/lucene/LuceneIterableTest.java
Tue Apr 9 16:37:44 2013
@@ -163,7 +163,7 @@ public final class LuceneIterableTest ex
RAMDirectory directory,
boolean createNew,
int startingId) throws
IOException {
- IndexWriter writer = new IndexWriter( directory, new
IndexWriterConfig(Version.LUCENE_41,new StandardAnalyzer(Version.LUCENE_41)));
+ IndexWriter writer = new IndexWriter( directory, new
IndexWriterConfig(Version.LUCENE_42,new StandardAnalyzer(Version.LUCENE_42)));
try {
for (int i = 0; i < DOCS.length; i++) {
Modified: mahout/trunk/pom.xml
URL:
http://svn.apache.org/viewvc/mahout/trunk/pom.xml?rev=1466110&r1=1466109&r2=1466110&view=diff
==============================================================================
--- mahout/trunk/pom.xml (original)
+++ mahout/trunk/pom.xml Tue Apr 9 16:37:44 2013
@@ -98,7 +98,7 @@
<maven.clover.multiproject>true</maven.clover.multiproject>
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
<hadoop.version>1.1.2</hadoop.version>
- <lucene.version>4.2.0</lucene.version>
+ <lucene.version>4.2.1</lucene.version>
</properties>
<issueManagement>
<system>Jira</system>
