bug-gnubg
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Bug-gnubg] MET interpolation algorithm


From: Joern Thyssen
Subject: Re: [Bug-gnubg] MET interpolation algorithm
Date: Thu, 4 Sep 2003 17:38:02 +0000
User-agent: Mutt/1.4.1i

On Thu, Sep 04, 2003 at 07:22:22PM +0200, Holger wrote
> At 16:13 04.09.2003 +0000, Joern Thyssen wrote:
> >On Wed, Sep 03, 2003 at 11:59:48PM +0200, Holger wrote
> >> gnubg (GUI) segfaulted when showing the MET right after startup with the
> >> new mec table. It did this whenever it should show interpolated values 
> >for
> >> this MET.
> >>
> >> I've put in a hack that should avoid gnubg crashing. But the 
> >interpolation
> >> algorithm seems to be faulty.
> >
> >No, the mec26.xml is faulty! There is only 24 post crawford equities
> >where there should be 25.
> >
> >When the extension algorithm start working on element 25 which contains
> >garbage, you get garbage for the remaining post crawford value.
> 
> Ah, I didn't look at this. Funny thing is that if e.g. Zadeh was loaded 
> before, everything seemed to be fine. I guess that the old values just 
> remained in place.

Yup, I guess so.

I've added some asserts in the code, so gnubg dies at the proper place.

If anyone has some time, it'll would probably be a good idea to add some
graceful exists instead.

Jørn

Attachment: pgpdBySTFfZx0.pgp
Description: PGP signature


reply via email to

[Prev in Thread] Current Thread [Next in Thread]