Luchen's code to dump out docids of docs in the index.

README.md

@@ -110,3 +110,13 @@ sh target/appassembler/bin/TweetSearcher -index twitter-index
 ```
 
 The demo starts up an HTTP server on port `8080`, but this can be changed with the `-port` option. Query via a web browser at `http://localhost:8080/search?query=query`. Try `birthday`, as there are always birthdays being celebrated.
+
+### IndexCounter:
+
+Output all the document IDs in a Lucene Index.
+
+```sh
+sh target/appassembler/bin/IndexCounter -indexPath /path/to/index \
+-docIdPath /path/to/save/docIds
+```
+=======

pom.xml

@@ -61,6 +61,10 @@
               <mainClass>io.anserini.nrts.TweetSearcher</mainClass>
               <name>TweetSearcher</name>
             </program>
+	    <program>
+              <mainClass>io.anserini.util.IndexCounter</mainClass>
+              <name>IndexCounter</name>
+            </program>
           </programs>
         </configuration>
       </plugin>

src/main/java/io/anserini/util/IndexCounter.java

@@ -0,0 +1,51 @@
+package io.anserini;
+
+import java.io.File;
+import java.io.IOException;
+import java.io.BufferedWriter;
+import java.io.FileWriter;
+
+import org.apache.lucene.index.DirectoryReader;
+import org.apache.lucene.store.FSDirectory;
+
+public class IndexCounter {
+
+	public void readIndex(String indexDir, String docIdPath) throws IOException{
+		FSDirectory dir = FSDirectory.open(new File(indexDir).toPath());
+		DirectoryReader reader = DirectoryReader.open(dir);
+
+		FileWriter fw = new FileWriter(new File(docIdPath));
+		BufferedWriter bw = new BufferedWriter(fw);
+		int len = reader.numDocs();
+		for (int i = 0; i < len; i ++){
+			String docName = reader.document(i).get("docname");
+			bw.write(docName + "\n");
+			//System.out.println("IndexCounter: " + i + " docs got");
+			if ((i & 65535) == 0){
+				System.out.println("IndexCounter: " + i + " docs got");
+			}
+			//System.out.println(docName);
+
+
+		}
+		bw.close();
+	}
+
+	public static void main(String[] clArgs) {
+		Args args = new Args(clArgs);
+		final String indexDir = args.getString("-indexPath") + "/index";
+		final String docIdPath = args.getString("-docIdPath");
+
+		args.check();
+
+		System.out.println("Index path: " + indexDir);
+		System.out.println("DocId path: " + docIdPath);
+		final IndexCounter ic = new IndexCounter();
+		try {
+			ic.readIndex(indexDir, docIdPath);
+		}
+		catch (IOException e){
+			e.printStackTrace();
+		}
+	}
+}