Message:
The following issue has been resolved as FIXED.
Resolver: David Bertoni
Date: Sun, 24 Oct 2004 12:02 PM
A fix is in CVS. Can you please verify, if possible? Thanks!
---------------------------------------------------------------------
View the issue:
http://issues.apache.org/jira/browse/XALANC-425
Here is an overview of the issue:
---------------------------------------------------------------------
Key: XALANC-425
Summary: The key() function can return wrong results on tree-fragments (memory
management bug)
Type: Bug
Status: Resolved
Resolution: FIXED
Project: XalanC
Components:
XalanC
Fix Fors:
CurrentCVS
Versions:
1.8
Assignee:
Reporter: Alain Le Guennec
Created: Wed, 30 Jun 2004 8:22 AM
Updated: Sun, 24 Oct 2004 12:02 PM
Environment: Operating System: Windows NT/2K
Platform: PC
Description:
I think there is a bug in key maps management in XalanC 1.8.0.
A new map is created when the key() function is used on a tree-fragment
to store key-value/node-list pairs for that key within that tree-fragment.
However, this new map is not deleted when the tree-fragment goes out of scope.
Moreover, the tree-fragment allocator might reuse the address of a previous
tree-fragment for a new one.
This results in the previous map(s) being reused for the new tree-fragment too,
instead of rebuilding new, up-to-date, map(s).
Below is a small test case (which does not depend on the actual input file),
and a proposed fix to be applied to function returnXResultTreeFrag() in
StylesheetExecutionContextDefault.cpp (which adds clean-up of the map entries
for the tree-fragment going out of scope).
Best regards,
Alain Le Guennec.
- -----------------------
- The test case:
- -----------------------
<?xml version="1.0" encoding = "ISO-8859-1"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform";
xmlns:exsl="http://exslt.org/common";
xmlns:math="http://exslt.org/math";
xmlns:dyn="http://exslt.org/dynamic";
xmlns:set="http://exslt.org/sets";
xmlns:xalan="http://xml.apache.org/xalan";
exclude-result-prefixes="exsl">
<xsl:key name="node-name" match="NODE" use="@name"/>
<xsl:template match="/">
<xsl:call-template name="bug"><xsl:with-param name="para"
select="'foo'"/></xsl:call-template>
<xsl:call-template name="bug"><xsl:with-param name="para"
select="'bar'"/></xsl:call-template>
</xsl:template>
<xsl:template name="bug">
<xsl:param name="para"/>
<xsl:variable name="fragment">
<NODE name="{$para}"/>
</xsl:variable>
<xsl:for-each select="exsl:node-set($fragment)/*">
<xsl:for-each select=".">
<xsl:variable name="homonymous" select="key('node-name',
@name)"/>
<xsl:if test="count($homonymous) = 0">
<xsl:message>Bug in key(): No NODE with
name=<xsl:value-of select="@name"/>
found!</xsl:message>
</xsl:if>
</xsl:for-each>
</xsl:for-each>
</xsl:template>
</xsl:stylesheet>
- -----------------------
And now the proposed fix:
- -----------------------
bool
StylesheetExecutionContextDefault::returnXResultTreeFrag(XResultTreeFrag*
theXResultTreeFrag)
{
assert(theXResultTreeFrag != 0);
KeyTablesTableType::iterator i =
m_keyTables.find(&theXResultTreeFrag->rtree());
if (i != m_keyTables.end())
{
delete (*i).second;
m_keyTables.erase(i);
}
if (m_xresultTreeFragAllocator.ownsObject(theXResultTreeFrag) == false)
{
return false;
}
else
{
XalanDocumentFragment* const theDocumentFragment =
theXResultTreeFrag->release();
m_xresultTreeFragAllocator.destroy(theXResultTreeFrag);
if (m_usePerInstanceDocumentFactory == true)
{
#if defined(XALAN_OLD_STYLE_CASTS)
m_documentAllocator.destroy((XalanSourceTreeDocument*)theDocumentFragment->getOwnerDocument());
#else
m_documentAllocator.destroy(static_cast<XalanSourceTreeDocument*>(theDocumentFragment->getOwnerDocument()));
#endif
}
#if defined(XALAN_OLD_STYLE_CASTS)
m_documentFragmentAllocator.destroy((XalanSourceTreeDocumentFragment*)theDocumentFragment);
#else
m_documentFragmentAllocator.destroy(static_cast<XalanSourceTreeDocumentFragment*>(theDocumentFragment));
#endif
return true;
}
}
---------------------------------------------------------------------
JIRA INFORMATION:
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
If you want more information on JIRA, or have a bug to report see:
http://www.atlassian.com/software/jira
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
