Bug 22155 - Support XPath queries that return objects other than node sets.
Summary: Support XPath queries that return objects other than node sets.
Status: CLOSED FIXED
Alias: None
Product: Xindice
Classification: Unclassified
Component: XPath (show other bugs)
Version: cvs head (1.1)
Hardware: Other other
: P3 enhancement
Target Milestone: ---
Assignee: Xindice Developers
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2003-08-06 04:04 UTC by Kevin O'Neill
Modified: 2004-11-16 19:05 UTC (History)
1 user (show)


Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Kevin O'Neill 2003-08-06 04:04:32 UTC 
Any xpath expression that returns anything but a nodeset will fail.
For example these three queries all result in exception.

boolean(//person)
string(//person[first='John' and last='Smith']/first)
count(//person[first='John' and last='Smith'])

To support this the we will need to update the NodeSet implementation to support
other datatypes.

I've added unit tests to the test suite that expose this problem
Comment 1 Vadim Gritsenko 2004-03-04 13:23:34 UTC 
Fixed, Closed.
Comment 2 Yogesh 2004-03-05 22:51:25 UTC 
Problem in XPathQueryResolver.ResultSet.prepareNextNode() when iterating over 
Collection and empty document i.e. '<?xml version="1.0" encoding="UTF-8"?>' is 
returned. Then the node returned by the document is null and a an exception is 
thrown at xp.execute(xpc, n, pfx) -- (exception attached at end). Checking for 
node != null fixes this problem. I'm not sure though how the empty document go 
into the DB in the first place.

DBDocument d = (DBDocument) context.getDocument(keySet[keyPos++]);
// dom2str(d) == "<?xml version="1.0" encoding="UTF-8"?>"
if (d == null) {
  continue;
}
Node n = d.getDocumentElement();
// n == null
if(n == null) continue; // added to fix the problem


Exception thrown by xp.execute(xpc, n, pfx):
java.lang.IllegalArgumentException: Node must be non-null for 
getDTMHandleFromNode
	at org.apache.xml.dtm.ref.DTMManagerDefault.getDTMHandleFromNode
(DTMManagerDefault.java:543)
	at org.apache.xpath.XPathContext.getDTMHandleFromNode
(XPathContext.java:220)
	at org.apache.xpath.XPath.execute(XPath.java:274)
	at 
org.apache.xindice.core.query.XPathQueryResolver$ResultSet.prepareNextNode
(XPathQueryResolver.java:1211)
	at 
org.apache.xindice.core.query.XPathQueryResolver$ResultSet.getNextNode
(XPathQueryResolver.java:1267)
	at org.apache.xindice.core.query.QueryUtil.queryResultsToDOM
(QueryUtil.java:75)
	at org.apache.xindice.core.query.QueryUtil.queryResultsToDOM
(QueryUtil.java:134)
	at org.apache.xindice.client.xmldb.embed.CollectionImpl.query
(CollectionImpl.java:437)
	at org.apache.xindice.client.xmldb.XindiceCollection.query
(XindiceCollection.java:259)
	at org.apache.xindice.client.xmldb.services.QueryService.query
(QueryService.java:90)
	at edu.indiana.extreme.gsx.util.XindiceUtil.executeXPath
(XindiceUtil.java:514)

Thanks.
--Yogesh
Comment 3 Vadim Gritsenko 2004-03-05 23:39:24 UTC 
Thanks for your observation. Can you also turn on DEBUG and see may be there are
some ignored exceptions preceding this?
Comment 4 Vadim Gritsenko 2004-03-20 13:48:55 UTC 
Yogesh, any updates?
Comment 5 Vadim Gritsenko 2004-05-21 12:48:21 UTC 
Added test for null getDocumentElement() with log statement.