This is the mail archive of the xconq7@sources.redhat.com mailing list for the Xconq project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: (remove) doesn't work


Hi Matthew,

On Tue, 2004-05-18 at 16:12, mskala@ansuz.sooke.bc.ca wrote:
> As far as I can tell, the (remove) form simply returns its second argument
> instead of doing its job.  Test case attached - this will fail regardless
> because of the lack of terrain types, but it'll also give an error about
> trying to match a length-7 list with another list that should also be
> length-7 but is in fact length-10.  The only shipped game module using
> (remove) is tailhook.g.  I haven't figured out whether (remove) is
> misbehaving in that file, but it definitely misbehaves in my test case.

I would definitely agree that 'remove' is broken, at least relative to
its documentation. I too tried to use it once, but it was not giving the
desired behavior, so I took a less convenient route that went around it.
I didn't mention anything at the time, because I thought it was just
some aspect of lispy behavior that I didn't fully comprehend (perhaps
this business about only being able to remove atoms rather than lists).

So, I guess the questions are, should it be able to remove lists? (It
would certainly be nice, since that is the way most people, myself
included, would want to use it) And, if so, how difficult would that
functionality be to add into the GDL parser? Hopefully someone will have
some insight into these questions, because otherwise I am going to have
to do some more "context gathering".

  Regards,
    Eric


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]