Author: kkrugler
Date: Thu Aug 12 22:54:57 2010
New Revision: 985028
URL: http://svn.apache.org/viewvc?rev=985028&view=rev
Log:
TIKA-478: Fix up missing end </body> and </html> tags for document with no real
content.
Modified:
tika/trunk/tika-core/src/main/java/org/apache/tika/sax/XHTMLContentHandler.java
tika/trunk/tika-parsers/src/test/java/org/apache/tika/parser/html/HtmlParserTest.java
Modified:
tika/trunk/tika-core/src/main/java/org/apache/tika/sax/XHTMLContentHandler.java
URL:
http://svn.apache.org/viewvc/tika/trunk/tika-core/src/main/java/org/apache/tika/sax/XHTMLContentHandler.java?rev=985028&r1=985027&r2=985028&view=diff
==============================================================================
---
tika/trunk/tika-core/src/main/java/org/apache/tika/sax/XHTMLContentHandler.java
(original)
+++
tika/trunk/tika-core/src/main/java/org/apache/tika/sax/XHTMLContentHandler.java
Thu Aug 12 22:54:57 2010
@@ -188,8 +188,9 @@ public class XHTMLContentHandler extends
public void endDocument() throws SAXException {
lazyEndHead();
- endElement("body");
- endElement("html");
+ super.endElement(XHTML, "body", "body");
+ super.endElement(XHTML, "html", "html");
+
endPrefixMapping("");
super.endDocument();
}
Modified:
tika/trunk/tika-parsers/src/test/java/org/apache/tika/parser/html/HtmlParserTest.java
URL:
http://svn.apache.org/viewvc/tika/trunk/tika-parsers/src/test/java/org/apache/tika/parser/html/HtmlParserTest.java?rev=985028&r1=985027&r2=985028&view=diff
==============================================================================
---
tika/trunk/tika-parsers/src/test/java/org/apache/tika/parser/html/HtmlParserTest.java
(original)
+++
tika/trunk/tika-parsers/src/test/java/org/apache/tika/parser/html/HtmlParserTest.java
Thu Aug 12 22:54:57 2010
@@ -444,7 +444,12 @@ public class HtmlParserTest extends Test
// link element should be in <head> section
assertTrue(Pattern.matches("(?s)<html.*<head>.*<link .*</head>.*$",
result));
+
+ // There should be ending elements.
+ assertTrue(Pattern.matches("(?s).*</body>.*</html>$", result));
+
}
+
}