fix a lucene bug

webmagic-lucene/pom.xml

@@ -24,7 +24,7 @@
         </dependency>
         <dependency>
             <groupId>us.codecraft</groupId>
-            <artifactId>webmagic-core</artifactId>
+            <artifactId>webmagic-extension</artifactId>
             <version>${project.version}</version>
         </dependency>
         <dependency>

webmagic-lucene/src/main/java/us/codecraft/webmagic/pipeline/LucenePipeline.java

@@ -33,16 +33,14 @@ public class LucenePipeline implements Pipeline {
 
     private Directory directory;
 
-    private IndexWriter indexWriter;
-
     private Analyzer analyzer;
 
+    private IndexWriterConfig config;
+
     private void init() throws IOException {
         analyzer = new StandardAnalyzer(Version.LUCENE_44);
         directory = new RAMDirectory();
-        IndexWriterConfig config = new IndexWriterConfig(Version.LUCENE_44, analyzer);
+        config = new IndexWriterConfig(Version.LUCENE_44, analyzer);
-        indexWriter = new IndexWriter(directory, config);
-        indexWriter.close();
     }
 
     public LucenePipeline() {
@@ -67,7 +65,6 @@ public class LucenePipeline implements Pipeline {
             documents.add(hitDoc);
         }
         ireader.close();
-        directory.close();
         return documents;
     }
 
@@ -85,7 +82,9 @@ public class LucenePipeline implements Pipeline {
             doc.add(new Field(objectEntry.getKey(), objectEntry.getValue().toString(), TextField.TYPE_STORED));
         }
         try {
+            IndexWriter indexWriter = new IndexWriter(directory, config);
             indexWriter.addDocument(doc);
+            indexWriter.close();
         } catch (IOException e) {
             e.printStackTrace();
         }

webmagic-lucene/src/main/test/java/us/codecraft/webmagic/lucene/OschinaBlog.java

@@ -0,0 +1,61 @@
+package us.codecraft.webmagic.lucene;
+
+import org.apache.lucene.document.Document;
+import org.apache.lucene.queryparser.classic.ParseException;
+import us.codecraft.webmagic.Site;
+import us.codecraft.webmagic.model.ExtractBy;
+import us.codecraft.webmagic.model.OOSpider;
+import us.codecraft.webmagic.model.TargetUrl;
+import us.codecraft.webmagic.pipeline.LucenePipeline;
+
+import java.io.IOException;
+import java.util.List;
+
+/**
+ * @author code4crafter@gmail.com <br>
+ * @date: 13-8-2 <br>
+ * Time: 上午7:52 <br>
+ */
+@TargetUrl("http://my.oschina.net/flashsword/blog/\\d+")
+public class OschinaBlog {
+
+    @ExtractBy("//title")
+    private String title;
+
+    @ExtractBy(value = "div.BlogContent", type = ExtractBy.Type.Css)
+    private String content;
+
+    @Override
+    public String toString() {
+        return "OschinaBlog{" +
+                "title='" + title + '\'' +
+                ", content='" + content + '\'' +
+                '}';
+    }
+
+    public static void main(String[] args) {
+        LucenePipeline pipeline = new LucenePipeline();
+        OOSpider.create(Site.me().addStartUrl("http://my.oschina.net/flashsword/blog"), OschinaBlog.class).pipeline(pipeline).runAsync();
+        while (true) {
+            try {
+                List<Document> search = pipeline.search("title", "webmagic");
+                System.out.println(search);
+                Thread.sleep(3000);
+            } catch (IOException e) {
+                e.printStackTrace();
+            } catch (ParseException e) {
+                e.printStackTrace();
+            } catch (InterruptedException e) {
+                e.printStackTrace();
+            }
+        }
+    }
+
+    public String getTitle() {
+        return title;
+    }
+
+    public String getContent() {
+        return content;
+    }
+}