You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@lucene.apache.org by rm...@apache.org on 2012/07/30 16:16:35 UTC
svn commit: r1367101 - in /lucene/dev/branches/branch_4x: ./ lucene/
lucene/analysis/
lucene/analysis/common/src/java/org/apache/lucene/analysis/ca/
lucene/analysis/common/src/java/org/apache/lucene/analysis/fr/
lucene/analysis/common/src/java/org/apac...
Author: rmuir
Date: Mon Jul 30 14:16:34 2012
New Revision: 1367101
URL: http://svn.apache.org/viewvc?rev=1367101&view=rev
Log:
LUCENE-3884: Move ElisionFilter out of .fr package
Added:
lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/util/ElisionFilter.java
- copied unchanged from r1367096, lucene/dev/trunk/lucene/analysis/common/src/java/org/apache/lucene/analysis/util/ElisionFilter.java
lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/util/ElisionFilterFactory.java
- copied unchanged from r1367096, lucene/dev/trunk/lucene/analysis/common/src/java/org/apache/lucene/analysis/util/ElisionFilterFactory.java
lucene/dev/branches/branch_4x/lucene/analysis/common/src/test/org/apache/lucene/analysis/util/TestElision.java
- copied unchanged from r1367096, lucene/dev/trunk/lucene/analysis/common/src/test/org/apache/lucene/analysis/util/TestElision.java
lucene/dev/branches/branch_4x/lucene/analysis/common/src/test/org/apache/lucene/analysis/util/TestElisionFilterFactory.java
- copied unchanged from r1367096, lucene/dev/trunk/lucene/analysis/common/src/test/org/apache/lucene/analysis/util/TestElisionFilterFactory.java
lucene/dev/branches/branch_4x/lucene/analysis/common/src/test/org/apache/lucene/analysis/util/frenchArticles.txt
- copied unchanged from r1367096, lucene/dev/trunk/lucene/analysis/common/src/test/org/apache/lucene/analysis/util/frenchArticles.txt
Removed:
lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/fr/ElisionFilter.java
lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/fr/ElisionFilterFactory.java
lucene/dev/branches/branch_4x/lucene/analysis/common/src/test/org/apache/lucene/analysis/fr/TestElision.java
lucene/dev/branches/branch_4x/lucene/analysis/common/src/test/org/apache/lucene/analysis/fr/TestElisionFilterFactory.java
lucene/dev/branches/branch_4x/lucene/analysis/common/src/test/org/apache/lucene/analysis/fr/frenchArticles.txt
Modified:
lucene/dev/branches/branch_4x/ (props changed)
lucene/dev/branches/branch_4x/lucene/ (props changed)
lucene/dev/branches/branch_4x/lucene/CHANGES.txt (contents, props changed)
lucene/dev/branches/branch_4x/lucene/analysis/ (props changed)
lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/ca/CatalanAnalyzer.java
lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/fr/FrenchAnalyzer.java
lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/ga/IrishAnalyzer.java
lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/it/ItalianAnalyzer.java
lucene/dev/branches/branch_4x/lucene/analysis/common/src/resources/META-INF/services/org.apache.lucene.analysis.util.TokenFilterFactory
Modified: lucene/dev/branches/branch_4x/lucene/CHANGES.txt
URL: http://svn.apache.org/viewvc/lucene/dev/branches/branch_4x/lucene/CHANGES.txt?rev=1367101&r1=1367100&r2=1367101&view=diff
==============================================================================
--- lucene/dev/branches/branch_4x/lucene/CHANGES.txt (original)
+++ lucene/dev/branches/branch_4x/lucene/CHANGES.txt Mon Jul 30 14:16:34 2012
@@ -68,6 +68,9 @@ API Changes
* LUCENE-3747: Support Unicode 6.1.0. (Steve Rowe)
+* LUCENE-3884: Moved ElisionFilter out of org.apache.lucene.analysis.fr
+ package into org.apache.lucene.analysis.util. (Robert Muir)
+
Optimizations
* LUCENE-4171: Performance improvements to Packed64.
Modified: lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/ca/CatalanAnalyzer.java
URL: http://svn.apache.org/viewvc/lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/ca/CatalanAnalyzer.java?rev=1367101&r1=1367100&r2=1367101&view=diff
==============================================================================
--- lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/ca/CatalanAnalyzer.java (original)
+++ lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/ca/CatalanAnalyzer.java Mon Jul 30 14:16:34 2012
@@ -24,7 +24,6 @@ import java.util.Arrays;
import org.apache.lucene.analysis.Analyzer;
import org.apache.lucene.analysis.core.LowerCaseFilter;
import org.apache.lucene.analysis.core.StopFilter;
-import org.apache.lucene.analysis.fr.ElisionFilter;
import org.apache.lucene.analysis.miscellaneous.KeywordMarkerFilter;
import org.apache.lucene.analysis.TokenStream;
import org.apache.lucene.analysis.Tokenizer;
@@ -32,6 +31,7 @@ import org.apache.lucene.analysis.snowba
import org.apache.lucene.analysis.standard.StandardFilter;
import org.apache.lucene.analysis.standard.StandardTokenizer;
import org.apache.lucene.analysis.util.CharArraySet;
+import org.apache.lucene.analysis.util.ElisionFilter;
import org.apache.lucene.analysis.util.StopwordAnalyzerBase;
import org.apache.lucene.util.Version;
import org.tartarus.snowball.ext.CatalanStemmer;
@@ -136,7 +136,7 @@ public final class CatalanAnalyzer exten
final Tokenizer source = new StandardTokenizer(matchVersion, reader);
TokenStream result = new StandardFilter(matchVersion, source);
if (matchVersion.onOrAfter(Version.LUCENE_36)) {
- result = new ElisionFilter(matchVersion, result, DEFAULT_ARTICLES);
+ result = new ElisionFilter(result, DEFAULT_ARTICLES);
}
result = new LowerCaseFilter(matchVersion, result);
result = new StopFilter(matchVersion, result, stopwords);
Modified: lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/fr/FrenchAnalyzer.java
URL: http://svn.apache.org/viewvc/lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/fr/FrenchAnalyzer.java?rev=1367101&r1=1367100&r2=1367101&view=diff
==============================================================================
--- lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/fr/FrenchAnalyzer.java (original)
+++ lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/fr/FrenchAnalyzer.java Mon Jul 30 14:16:34 2012
@@ -28,6 +28,7 @@ import org.apache.lucene.analysis.standa
import org.apache.lucene.analysis.standard.StandardTokenizer;
import org.apache.lucene.analysis.standard.StandardAnalyzer; // for javadoc
import org.apache.lucene.analysis.util.CharArraySet;
+import org.apache.lucene.analysis.util.ElisionFilter;
import org.apache.lucene.analysis.util.StopwordAnalyzerBase;
import org.apache.lucene.analysis.util.WordlistLoader;
import org.apache.lucene.util.IOUtils;
@@ -97,6 +98,11 @@ public final class FrenchAnalyzer extend
/** File containing default French stopwords. */
public final static String DEFAULT_STOPWORD_FILE = "french_stop.txt";
+ /** Default set of articles for ElisionFilter */
+ public static final CharArraySet DEFAULT_ARTICLES = CharArraySet.unmodifiableSet(
+ new CharArraySet(Version.LUCENE_CURRENT, Arrays.asList(
+ "l", "m", "t", "qu", "n", "s", "j"), true));
+
/**
* Contains words that should be indexed but not stemmed.
*/
@@ -185,7 +191,7 @@ public final class FrenchAnalyzer extend
if (matchVersion.onOrAfter(Version.LUCENE_31)) {
final Tokenizer source = new StandardTokenizer(matchVersion, reader);
TokenStream result = new StandardFilter(matchVersion, source);
- result = new ElisionFilter(matchVersion, result);
+ result = new ElisionFilter(result, DEFAULT_ARTICLES);
result = new LowerCaseFilter(matchVersion, result);
result = new StopFilter(matchVersion, result, stopwords);
if(!excltable.isEmpty())
Modified: lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/ga/IrishAnalyzer.java
URL: http://svn.apache.org/viewvc/lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/ga/IrishAnalyzer.java?rev=1367101&r1=1367100&r2=1367101&view=diff
==============================================================================
--- lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/ga/IrishAnalyzer.java (original)
+++ lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/ga/IrishAnalyzer.java Mon Jul 30 14:16:34 2012
@@ -23,7 +23,6 @@ import java.util.Arrays;
import org.apache.lucene.analysis.Analyzer;
import org.apache.lucene.analysis.core.StopFilter;
-import org.apache.lucene.analysis.fr.ElisionFilter;
import org.apache.lucene.analysis.miscellaneous.KeywordMarkerFilter;
import org.apache.lucene.analysis.TokenStream;
import org.apache.lucene.analysis.Tokenizer;
@@ -31,6 +30,7 @@ import org.apache.lucene.analysis.snowba
import org.apache.lucene.analysis.standard.StandardFilter;
import org.apache.lucene.analysis.standard.StandardTokenizer;
import org.apache.lucene.analysis.util.CharArraySet;
+import org.apache.lucene.analysis.util.ElisionFilter;
import org.apache.lucene.analysis.util.StopwordAnalyzerBase;
import org.apache.lucene.util.Version;
import org.tartarus.snowball.ext.IrishStemmer;
@@ -140,7 +140,7 @@ public final class IrishAnalyzer extends
StopFilter s = new StopFilter(matchVersion, result, HYPHENATIONS);
s.setEnablePositionIncrements(false);
result = s;
- result = new ElisionFilter(matchVersion, result, DEFAULT_ARTICLES);
+ result = new ElisionFilter(result, DEFAULT_ARTICLES);
result = new IrishLowerCaseFilter(result);
result = new StopFilter(matchVersion, result, stopwords);
if(!stemExclusionSet.isEmpty())
Modified: lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/it/ItalianAnalyzer.java
URL: http://svn.apache.org/viewvc/lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/it/ItalianAnalyzer.java?rev=1367101&r1=1367100&r2=1367101&view=diff
==============================================================================
--- lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/it/ItalianAnalyzer.java (original)
+++ lucene/dev/branches/branch_4x/lucene/analysis/common/src/java/org/apache/lucene/analysis/it/ItalianAnalyzer.java Mon Jul 30 14:16:34 2012
@@ -24,7 +24,6 @@ import java.util.Arrays;
import org.apache.lucene.analysis.Analyzer;
import org.apache.lucene.analysis.core.LowerCaseFilter;
import org.apache.lucene.analysis.core.StopFilter;
-import org.apache.lucene.analysis.fr.ElisionFilter;
import org.apache.lucene.analysis.miscellaneous.KeywordMarkerFilter;
import org.apache.lucene.analysis.TokenStream;
import org.apache.lucene.analysis.Tokenizer;
@@ -32,6 +31,7 @@ import org.apache.lucene.analysis.snowba
import org.apache.lucene.analysis.standard.StandardFilter;
import org.apache.lucene.analysis.standard.StandardTokenizer;
import org.apache.lucene.analysis.util.CharArraySet;
+import org.apache.lucene.analysis.util.ElisionFilter;
import org.apache.lucene.analysis.util.StopwordAnalyzerBase;
import org.apache.lucene.analysis.util.WordlistLoader;
import org.apache.lucene.util.IOUtils;
@@ -140,7 +140,7 @@ public final class ItalianAnalyzer exten
final Tokenizer source = new StandardTokenizer(matchVersion, reader);
TokenStream result = new StandardFilter(matchVersion, source);
if (matchVersion.onOrAfter(Version.LUCENE_32)) {
- result = new ElisionFilter(matchVersion, result, DEFAULT_ARTICLES);
+ result = new ElisionFilter(result, DEFAULT_ARTICLES);
}
result = new LowerCaseFilter(matchVersion, result);
result = new StopFilter(matchVersion, result, stopwords);
Modified: lucene/dev/branches/branch_4x/lucene/analysis/common/src/resources/META-INF/services/org.apache.lucene.analysis.util.TokenFilterFactory
URL: http://svn.apache.org/viewvc/lucene/dev/branches/branch_4x/lucene/analysis/common/src/resources/META-INF/services/org.apache.lucene.analysis.util.TokenFilterFactory?rev=1367101&r1=1367100&r2=1367101&view=diff
==============================================================================
--- lucene/dev/branches/branch_4x/lucene/analysis/common/src/resources/META-INF/services/org.apache.lucene.analysis.util.TokenFilterFactory (original)
+++ lucene/dev/branches/branch_4x/lucene/analysis/common/src/resources/META-INF/services/org.apache.lucene.analysis.util.TokenFilterFactory Mon Jul 30 14:16:34 2012
@@ -41,7 +41,6 @@ org.apache.lucene.analysis.en.PorterStem
org.apache.lucene.analysis.es.SpanishLightStemFilterFactory
org.apache.lucene.analysis.fa.PersianNormalizationFilterFactory
org.apache.lucene.analysis.fi.FinnishLightStemFilterFactory
-org.apache.lucene.analysis.fr.ElisionFilterFactory
org.apache.lucene.analysis.fr.FrenchLightStemFilterFactory
org.apache.lucene.analysis.fr.FrenchMinimalStemFilterFactory
org.apache.lucene.analysis.ga.IrishLowerCaseFilterFactory
@@ -89,4 +88,5 @@ org.apache.lucene.analysis.sv.SwedishLig
org.apache.lucene.analysis.synonym.SynonymFilterFactory
org.apache.lucene.analysis.th.ThaiWordFilterFactory
org.apache.lucene.analysis.tr.TurkishLowerCaseFilterFactory
+org.apache.lucene.analysis.util.ElisionFilterFactory
org.apache.lucene.collation.CollationKeyFilterFactory