[Archive-access-cvs] SF.net SVN: archive-access:[3211] tags/nutchwax-0_13-JIRA-WAX-75/archive/ src/plugin/index-nutchwax/src/java/org/archive/nutchwax/index/DateIndexer. java

[<bi...@us...>]

Revision: 3211
          http://archive-access.svn.sourceforge.net/archive-access/?rev=3211&view=rev
Author:   binzino
Date:     2010-08-09 23:51:25 +0000 (Mon, 09 Aug 2010)

Log Message:
-----------
Added some comments.

Modified Paths:
--------------
    tags/nutchwax-0_13-JIRA-WAX-75/archive/src/plugin/index-nutchwax/src/java/org/archive/nutchwax/index/DateIndexer.java

Modified: tags/nutchwax-0_13-JIRA-WAX-75/archive/src/plugin/index-nutchwax/src/java/org/archive/nutchwax/index/DateIndexer.java
===================================================================
--- tags/nutchwax-0_13-JIRA-WAX-75/archive/src/plugin/index-nutchwax/src/java/org/archive/nutchwax/index/DateIndexer.java	2010-08-09 16:04:53 UTC (rev 3210)
+++ tags/nutchwax-0_13-JIRA-WAX-75/archive/src/plugin/index-nutchwax/src/java/org/archive/nutchwax/index/DateIndexer.java	2010-08-09 23:51:25 UTC (rev 3211)
@@ -36,7 +36,16 @@
 import org.apache.nutch.parse.Parse;
 
 /** 
+ * Add fields to NutchDocument related to dates.
  *
+ * All of the crawl dates are taken from the Parse object, reduced to
+ * a set of unique dates (in case of redundancies), then the dates
+ * are added to the NutchDocument in multiple fields:
+ * <ul>
+ * <li><code>date</code> stored for later retrieval.</li>
+ * <li><code>year</code> only the year (YYYY), indexed not stored.</l>
+ * <li><code>yearmonth</code> year and month (YYYYMM), indexed not stored.</li>
+ * </ul>
  */
 public class DateIndexer implements IndexingFilter
 {
@@ -55,16 +64,14 @@
   }
 
   /**
-   * <p>Set Lucene document field to fixed value.</p>
-   * <p>
-   *   Remove field if specified value is <code>null</code>.
-   * </p>
+   *
    */
   public NutchDocument filter( NutchDocument doc, Parse parse, Text url, CrawlDatum datum, Inlinks inlinks )
     throws IndexingException
   {
     Metadata meta = parse.getData().getContentMeta();
 
+    // Use a Set to reduce list of dates to just unique values.
     Set<String> dates = new HashSet<String>( Arrays.asList( meta.getValues( "date" ) ) );
 
     for ( String date : dates )


This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.