Index: lucene/contrib/demo/src/test/org/apache/lucene/demo/test-files/html/test2.html
===================================================================
--- lucene/contrib/demo/src/test/org/apache/lucene/demo/test-files/html/test2.html (revision 0)
+++ lucene/contrib/demo/src/test/org/apache/lucene/demo/test-files/html/test2.html (revision 0)
@@ -0,0 +1,9 @@
+
+
+
+
+
+
+ This document is actually not about cats!
+
+
Property changes on: lucene\contrib\demo\src\test\org\apache\lucene\demo\test-files\html\test2.html
___________________________________________________________________
Added: svn:eol-style
+ native
Index: lucene/contrib/demo/src/test/org/apache/lucene/demo/test-files/queries2.txt
===================================================================
--- lucene/contrib/demo/src/test/org/apache/lucene/demo/test-files/queries2.txt (revision 0)
+++ lucene/contrib/demo/src/test/org/apache/lucene/demo/test-files/queries2.txt (revision 0)
@@ -0,0 +1 @@
++contents:dogs +contents:fish
Property changes on: lucene\contrib\demo\src\test\org\apache\lucene\demo\test-files\queries2.txt
___________________________________________________________________
Added: svn:eol-style
+ native
Index: lucene/contrib/demo/src/test/org/apache/lucene/demo/TestDemo.java
===================================================================
--- lucene/contrib/demo/src/test/org/apache/lucene/demo/TestDemo.java (revision 1031463)
+++ lucene/contrib/demo/src/test/org/apache/lucene/demo/TestDemo.java (working copy)
@@ -43,4 +43,24 @@
System.setOut(outSave);
}
}
+
+ // LUCENE-591
+ public void testIndexKeywords() throws Exception {
+ File dir = getDataFile("test-files/html");
+ File indexDir = new File(TEMP_DIR, "demoIndex2");
+ IndexHTML.main(new String[] { "-create", "-index", indexDir.getPath(), dir.getPath() });
+ File queries = getDataFile("test-files/queries2.txt");
+ PrintStream outSave = System.out;
+ try {
+ ByteArrayOutputStream bytes = new ByteArrayOutputStream();
+ PrintStream fakeSystemOut = new PrintStream(bytes);
+ System.setOut(fakeSystemOut);
+ SearchFiles.main(new String[] { "-index", indexDir.getPath(), "-queries", queries.getPath()});
+ fakeSystemOut.flush();
+ String output = bytes.toString(); // intentionally use default encoding
+ assertTrue(output.contains("1 total matching documents"));
+ } finally {
+ System.setOut(outSave);
+ }
+ }
}
Index: lucene/contrib/demo/src/java/org/apache/lucene/demo/HTMLDocument.java
===================================================================
--- lucene/contrib/demo/src/java/org/apache/lucene/demo/HTMLDocument.java (revision 1031463)
+++ lucene/contrib/demo/src/java/org/apache/lucene/demo/HTMLDocument.java (working copy)
@@ -70,6 +70,11 @@
// Add the tag-stripped contents as a Reader-valued Text field so it will
// get tokenized and indexed.
doc.add(new Field("contents", parser.getReader()));
+
+ // add any document keywords if they exist
+ String keywords = parser.getMetaTags().getProperty("keywords");
+ if (keywords != null)
+ doc.add(new Field("contents", keywords, Field.Store.NO, Field.Index.ANALYZED));
// Add the summary as a field that is stored and returned with
// hit documents for display.
Index: lucene/contrib/CHANGES.txt
===================================================================
--- lucene/contrib/CHANGES.txt (revision 1031467)
+++ lucene/contrib/CHANGES.txt (working copy)
@@ -143,6 +143,9 @@
* LUCENE-590: Demo HTML parser gives incorrect summaries when title is repeated as a heading
(Curtis d'Entremont via Robert Muir)
+
+* LUCENE-591: The demo indexer now indexes meta keywords.
+ (Curtis d'Entremont via Robert Muir)
API Changes