On 4/13/10 11:09 AM, Jaime Villate wrote:
> On Tue, 2010-04-13 at 09:43 -0400, Raymond Toy wrote:
>
>> I suspect the bug was caused in rev 1.148. In this change, xrange and
>> yrange were removed from the arg list of $plot3d, but xrange and
>> yrange
>> are still referenced in the body of $plot3d. Lisp warns about these
>> undefined variables. There might be some other undefined variables; I
>> can't remember and I don't have the compile log around.
>>
> Hi Ray,
> that must be the source of the problem. I removed xrange and yrange from
> the list of arguments, to allow for a more general syntax in which
> plot3d accepts more than one surface. The xrange and yrange are now two
> more options which are saved in a "property list" name features. When
> needed, they should be called as (getf :xrange features).
> If you saw xrange and yrange being used as variables, that is definitely
> an error. I will try to fix it. If you find any other undefined
> variables in your log file, please let me know.
>
Since you've fixed some of the issues with clisp, you must also have a
compiler warnings from clisp. But anyway, here are the warnings from
clisp. xrange and yrange are undefined in $plot3d. I think the problem
is that you changed the &aux variables xrange and yrange to rangex and
rangey. But rangex and rangey aren't used anywhere in $plot3d.
Renaming them to xrange/yrange fixes the warnings.
There is one more warning from clisp: the variable AR is undefined.
This happens in the code to handle the ZIC plot format. I don't think
zic has been supported for years, so maybe we just just remove that
code. (Does anyone use zic anymore?)
Ray