- From: r.britten at auckland.ac.nz (Randall Britten)
- Subject: [cellml-dev] FW: r2129 - CellML_DOM_API/trunk/CCGS/sources
- Date: Thu, 13 Mar 2008 18:02:12 +1300
Hi Justin
I would suggest adding a test to the unit-tests for this. I.e. a test that
fails prior to the patch, and passes after the patch.
Regards,
Randall
>
-----Original Message-----
>
From: automated-notifications-bounces at cellml.org [mailto:automated-
>
notifications-bounces at cellml.org] On Behalf Of CellML Automated
>
Notifications
>
Sent: Thursday, 13 March 2008 5:51 p.m.
>
To: automated-notifications at cellml.org
>
Subject: r2129 - CellML_DOM_API/trunk/CCGS/sources
>
>
Author: jmar150
>
Date: 2008-03-13 17:51:12 +1300 (Thu, 13 Mar 2008)
>
New Revision: 2129
>
>
Modified:
>
CellML_DOM_API/trunk/CCGS/sources/CCGSGenerator.cpp
>
Log:
>
This fixes a potential problem where, if a model was unsuitably
>
constrained, and had an
>
initial value assignment to a variable that depended upon some other
>
variable which was not
>
yet defined, a null pointer deref would occur, as there would be an
>
unused edge with NULL
>
mMaths, and we want to add ref all the maths of all the unused edges.
>
>
>
>
Modified: CellML_DOM_API/trunk/CCGS/sources/CCGSGenerator.cpp
>
===================================================================
>
--- CellML_DOM_API/trunk/CCGS/sources/CCGSGenerator.cpp 2008-03-12
>
22:40:54 UTC (rev 2128)
>
+++ CellML_DOM_API/trunk/CCGS/sources/CCGSGenerator.cpp 2008-03-13
>
04:51:12 UTC (rev 2129)
>
@@ -136,8 +136,13 @@
>
for (uel = mUnusedEdges.begin(); uel != mUnusedEdges.end(); uel++)
>
{
>
iface::mathml_dom::MathMLApplyElement* mae = (*uel)->mMaths;
>
- mae->add_ref();
>
- mCodeInfo->mFlaggedEquations.push_back(mae);
>
+ // If there was an initial value that was set to a variable
>
which could not
>
+ // be computed, it will be an unused edge with null mMaths.
>
+ if (mae != NULL)
>
+ {
>
+ mae->add_ref();
>
+ mCodeInfo->mFlaggedEquations.push_back(mae);
>
+ }
>
}
>
>
mCodeInfo->mConstraintLevel =
>
iface::cellml_services::UNSUITABLY_CONSTRAINED;
>
>
_______________________________________________
>
automated-notifications mailing list
>
automated-notifications at cellml.org
>
http://www.cellml.org/mailman/listinfo/automated-notifications
- [cellml-dev] FW: r2129 - CellML_DOM_API/trunk/CCGS/sources, Randall Britten, 03/13/2008
Archive powered by MHonArc 2.6.18.