[ http://issues.apache.org/jira/browse/XALANJ-1625?page=all ]
Brian Minchau updated XALANJ-1625:
----------------------------------
Version: Latest Development Code
(was: 2.5)
> org.apache.xml.dtm.ref.DTMDefaultBase (SuballocatedIntVector)-index-error
> --------------------------------------------------------------------------
>
> Key: XALANJ-1625
> URL: http://issues.apache.org/jira/browse/XALANJ-1625
> Project: XalanJ2
> Type: Bug
> Components: DTM
> Versions: Latest Development Code
> Environment: Operating System: Other
> Platform: Other
> Reporter: Oliver Haider
> Assignee: Xalan Developers Mailing List
>
> There is a misshap in the class bug in org.apache.xml.dtm.ref.DTMDefaultBase
> (at the function SuballocatedIntVector)
> In the last few lines the local var candiate is assigned
> (candidate=m_namespaceDeclSetElements.elementAt(--wouldBeAt);)
> the problem is the --wouldBeAt since it is checked for >= 0 thus it might be
> 0
> and then in the get-function -1.
> regards
> oliver haider
> protected SuballocatedIntVector findNamespaceContext(int elementNodeIndex)
> {
> if (null!=m_namespaceDeclSetElements)
> {
> // %OPT% Is binary-search really saving us a lot versus linear?
> // (... It may be, in large docs with many NS decls.)
> int wouldBeAt=findInSortedSuballocatedIntVector
> (m_namespaceDeclSetElements,
> elementNodeIndex);
> if(wouldBeAt>=0) // Found it
> return (SuballocatedIntVector) m_namespaceDeclSets.elementAt
> (wouldBeAt);
> if(wouldBeAt == -1) // -1-wouldbeat == 0
> return null; // Not after anything; definitely not found
> // Not found, but we know where it should have been.
> // Search back until we find an ancestor or run out.
> wouldBeAt=-1-wouldBeAt;
> // Decrement wouldBeAt to find last possible ancestor
> int candidate=m_namespaceDeclSetElements.elementAt(-- wouldBeAt);
> int ancestor=_parent(elementNodeIndex);
> // Special case: if the candidate is before the given node, and
> // is in the earliest possible position in the document, it
> // must have the namespace declarations we're interested in.
> if (wouldBeAt == 0 && candidate < ancestor) {
> int rootHandle = getDocumentRoot(makeNodeHandle(elementNodeIndex));
> int rootID = makeNodeIdentity(rootHandle);
> int uppermostNSCandidateID;
> if (getNodeType(rootHandle) == DTM.DOCUMENT_NODE) {
> int ch = _firstch(rootID);
> uppermostNSCandidateID = (ch != DTM.NULL) ? ch : rootID;
> } else {
> uppermostNSCandidateID = rootID;
> }
> if (candidate == uppermostNSCandidateID) {
> return (SuballocatedIntVector)m_namespaceDeclSets.elementAt
> (wouldBeAt);
> }
> }
> while(wouldBeAt>=0 && ancestor>0)
> {
> if (candidate==ancestor) {
> // Found ancestor in list
> return (SuballocatedIntVector)m_namespaceDeclSets.elementAt
> (wouldBeAt);
> } else if (candidate<ancestor) {
> // Too deep in tree
> do {
> ancestor=_parent(ancestor);
> } while (candidate < ancestor);
> } else {
> // Too late in list
> candidate=m_namespaceDeclSetElements.elementAt(--wouldBeAt);
> }
> }
> }
> return null; // No namespaces known at this node
> }
--
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators:
http://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see:
http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
