From 18dd9a5f183d95914791784f8e475ea5c1f7945d Mon Sep 17 00:00:00 2001
From: webdevuser <webdevuser@newwebdevserver-1.(none)>
Date: Thu, 19 May 2011 20:21:25 +0200
Subject: [PATCH] Add "articles" setting to specify a list of stopword for the
 ElisionFilter.

---
 .../index/analysis/Analysis.java              | 23 +++++++++++++++++++
 .../analysis/ElisionTokenFilterFactory.java   | 13 +++++++++--
 2 files changed, 34 insertions(+), 2 deletions(-)

diff --git a/modules/elasticsearch/src/main/java/org/elasticsearch/index/analysis/Analysis.java b/modules/elasticsearch/src/main/java/org/elasticsearch/index/analysis/Analysis.java
index d0b98062d3c..2f3b592ab42 100644
--- a/modules/elasticsearch/src/main/java/org/elasticsearch/index/analysis/Analysis.java
+++ b/modules/elasticsearch/src/main/java/org/elasticsearch/index/analysis/Analysis.java
@@ -118,6 +118,29 @@ public class Analysis {
             .put("_turkish_", TurkishAnalyzer.getDefaultStopSet())
             .immutableMap();
 
+    public static Set<?> parseArticles(Environment env, Settings settings) {
+        String value = settings.get("articles");
+        if (value != null) {
+            if ("_none_".equals(value)) {
+                return ImmutableSet.of();
+            } else {
+                return ImmutableSet.copyOf(Strings.commaDelimitedListToSet(value));
+            }
+        }
+        String[] articles = settings.getAsArray("articles", null);
+        if (articles != null) {
+            Set setArticles = new HashSet<String>(Arrays.asList(articles));
+            return setArticles;
+        }
+        Set<String> pathLoadedArticles = getWordSet(env, settings, "articles");
+        if (pathLoadedArticles != null) {
+            Set setArticles = new HashSet<String>(pathLoadedArticles);
+            return setArticles;
+        }
+
+        return null;
+    }
+
     public static Set<?> parseStopWords(Environment env, Settings settings, Set<?> defaultStopWords) {
         String value = settings.get("stopwords");
         if (value != null) {
diff --git a/modules/elasticsearch/src/main/java/org/elasticsearch/index/analysis/ElisionTokenFilterFactory.java b/modules/elasticsearch/src/main/java/org/elasticsearch/index/analysis/ElisionTokenFilterFactory.java
index 919b29c87a6..a01f12a4510 100644
--- a/modules/elasticsearch/src/main/java/org/elasticsearch/index/analysis/ElisionTokenFilterFactory.java
+++ b/modules/elasticsearch/src/main/java/org/elasticsearch/index/analysis/ElisionTokenFilterFactory.java
@@ -24,20 +24,29 @@ import org.apache.lucene.analysis.fr.ElisionFilter;
 import org.elasticsearch.common.inject.Inject;
 import org.elasticsearch.common.inject.assistedinject.Assisted;
 import org.elasticsearch.common.settings.Settings;
+import org.elasticsearch.env.Environment;
 import org.elasticsearch.index.Index;
 import org.elasticsearch.index.settings.IndexSettings;
 
+import java.util.Set;
 
 /**
  * @author kimchy (Shay Banon)
  */
 public class ElisionTokenFilterFactory extends AbstractTokenFilterFactory {
 
-    @Inject public ElisionTokenFilterFactory(Index index, @IndexSettings Settings indexSettings, @Assisted String name, @Assisted Settings settings) {
+    private final Set<?> articles;
+
+    @Inject public ElisionTokenFilterFactory(Index index, @IndexSettings Settings indexSettings, Environment env, @Assisted String name, @Assisted Settings settings) {
         super(index, indexSettings, name, settings);
+	this.articles = Analysis.parseArticles(env, settings);
     }
 
     @Override public TokenStream create(TokenStream tokenStream) {
-        return new ElisionFilter(version, tokenStream);
+	if (articles == null) {
+	    return new ElisionFilter(version, tokenStream);
+	} else {
+	    return new ElisionFilter(version, tokenStream, articles);
+	}
     }
 }
\ No newline at end of file