全文檢索筆記 - Lucene.Net (2) 盤古分詞

private static string IndexPath = "E:\\LuceneIndex";

public static void SimpleDemo()

{

    //指定索引資料儲存目錄

    var fsDir = FSDirectory.Open(IndexPath);

    //建立IndexWriter

    using (var idxWriter = new IndexWriter(

        fsDir, //儲存目錄

        new PanGuAnalyzer(), //使用盤古分詞器

        true, //清除原有索引，重新建立

        IndexWriter.MaxFieldLength.UNLIMITED //不限定欄位內容長度

        ))

    {

        //示範為兩份文件建立索引

        var doc = new Document();

        //每份文件有兩個Field: Source、Word

        doc.Add(new Field("Source", "阿甘正傳", Field.Store.YES, Field.Index.ANALYZED));

        doc.Add(new Field("Word", "人生就像一盒巧克力，你永遠也不會知道你將拿到什麼。", 

            Field.Store.YES, Field.Index.ANALYZED));

        idxWriter.AddDocument(doc);

        doc = new Document();

        doc.Add(new Field("Source", "Spider Man", Field.Store.YES, Field.Index.ANALYZED));

        doc.Add(new Field("Word", "Remember, with great power, comes great responsibility.", 

            Field.Store.YES,

            Field.Index.ANALYZED));

        idxWriter.AddDocument(doc);

        //建立索引

        idxWriter.Commit();

        idxWriter.Optimize();

    }

    //查詢示範

    //若不需刪除文件或修改Norms，第二個參數傳入true採唯讀方式效能較好

    var searcher = new IndexSearcher(fsDir, true); 

    //指定欄位名傳入參數

    QueryParser qp = new QueryParser(Version.LUCENE_30, "Word", new PanGuAnalyzer());

    Query q = qp.Parse("巧克力");

    var hits = searcher.Search(q, 10); //查詢前10筆

    Debug.WriteLine($"找到{hits.TotalHits}筆");

    foreach (var doc in hits.ScoreDocs)

    {

        Debug.WriteLine($"{searcher.Doc(doc.Doc).Get("Word")}");

    }

}