Compilation Errors

[Guest xionmark]

Greetings,

I've gotten an error when trying compile the ODE code, WIN-XP, H8.0.347. Here's the error message:

c:\Documents and Settings\mstory\My Documents\work\ODE\ode\SIM_SolverODE\SIM_Sol
verODE.h(52) : error C2259: 'SIM_SolverODE' : cannot instantiate abstract class
        due to following members:
        'SIM_Solver::SIM_Result SIM_Solver::solveObjectsSubclass(SIM_Engine &,SI
M_ObjectArray &,SIM_ObjectArray &,SIM_ObjectArray &,const SIM_Time &)' : pure virtual function was not defined
        C:/Program Files\Side Effects Software\Houdini 8.0.347\toolkit\include\SIM\SIM_Solver.h(120) : see declaration of 'SIM_Solver::solveObjectsSubclass'
c:\Program Files\Side Effects Software\Houdini 8.0.347\toolkit\include\SIM\SIM_GroupArray.h(26) : fatal error C1903: unable to recover from previous error(s); s
topping compilation

However, the solveObjectsSubclass() virtual function is in fact defined, haven't had a chance yet to look at the error related to SIM_GroupArray().

I'm only able to work on this at home, so my time is extremely limited, thus my posting to ask for hints on this. The ODE code I used is the latest on the sourceforge site.

--Mark

[Guest xionmark]

Hi there,

I've poured over the source code and this makes no sense why I'm getting this error, the solveObjectsSubclass() is defined (it's the guy doing most of the work!), could this be a .NET 2003 issue? I can't imagine that it would be.

Has anyone recently compiled this on a WIN box lately?

Unfortunately Charlex is almost 100% Windows for the artist's workstations, so I need to build it for WIN-XP.

This feels like one of those errors that's right under my nose, but I can't sniff it out yet ...

(And to complicate things, the only WIN box that hcustom works on for me is mine at home ... but that's the subject for a different post)

--Mark

[edward]

Maybe the parameters changed for the method? Carefully make sure that your subclass' function declaration matches that of the base class.

[Guest xionmark]

Maybe the parameters changed for the method? Carefully make sure that your subclass' function declaration matches that of the base class.

21132[/snapback]

Yep!

That's it!

There's now only 2 SIM_ObjectArray's being passed in.

virtual void solveObjectsSubclass (SIM_Engine &engine, SIM_ObjectArray &objects, SIM_ObjectArray &newobjects, const SIM_Time &timestep) const =0

Thanks Edward, I should built the Doxy HDK files for H8.0.347 a few days ago ... oh well!

Now to find the impications of the changes to the SIM_SolverODE class.

You're the best!

--Mark

[Guest xionmark]

Ok, cool, it looks like that SIM_ObjectArray isn't being used for anything currently.

I'll see if I can't figure out what's up with this WIN box I'm on, (hcustom keeps failing), then I won't have to wait to get home to compile! :-)

Thanks again!

--Mark

[Guest xionmark]

OK, I got hcustom working thanks to the tips from this thread:

http://odforce.net/forum/index.php?showtop...t=0entry20879

(for some reason the .NET "Command Prompt" still doesn't work on this machine, I'll figure that out later)

Anyway, I spoke too quickly in the last reply, I was looking at older H8 docs, the newest does on fact say:

virtual SIM_Result  solveObjectsSubclass (SIM_Engine &engine, SIM_ObjectArray &objects, SIM_ObjectArray &newobjects, SIM_ObjectArray &feedbacktoobjects, const SIM_Time &timestep)=0

But now I get a bunch of linking errors ... still plugging along ...

--Mark