Hello,
in my experience “Dr. Watson” was fairly useful because it allow at least to identify the exact point where the crash of the software happens.
The most part of mistakes that I have encountered are due to a code, that has been badly developed from our side, that in certain situations will carry out some non valid operations that cause the a.m. crash problems.
In this case (it never happened) instead is the function National "SetActivePanel()" that blocks itself without giving any error reply as stand out from the Dr. Watson log.
I am asking myself if there are some limits to its utilisation, or if it exists some other functions that could survey eventual errors, or perhaps to check that the panel is really active and the function could really be executed, or if it ‘ s possible to execute the same function "SetActivePanel()" in another way.
This is the summary of my code:
int CVICALLBACK newpanel_callback (int panel, int control, int event, void *callbackData, int eventData1, int eventData2)
{
switch (event)
{
case EVENT_COMMIT:
case EVENT_GOT_FOCUS:
.............................
.............................
some other code..............
creation of other windows...
.............................
.............................
// re-activation of the original panel
es= SetActivePanel(panel); <------ "point of the crash".
if(es < 0)
{
Actual spy text "Error SetActivePanel() number : ", es);
}
}
}
