[Gmsh] "For" loop with a variable incrementation step

[Christophe Geuzaine]

Lucas - I've committed a fix in the code, which should at least allow to 
bypass the issue.

Loops now correctly update the value of the loop variable before the 
next iteration. You can still not change "step" in your example (because 
the list "{0:100:step}" is fixed), but you can change the variable "k" 
inside the loop, e.g.

For k In {0:100:1}
   Printf("k = %f", k );
   k++;
EndFor



On 20/10/09 22:11, Christophe Geuzaine wrote:
> Lucas Blattner Martinho wrote:
>> Dear Gmsh developers,
>>
>>
>> I am trying to use Gmsh's "For" loop with a variable incrementation
>> step, however I am not obtaining the expected results.
>>
>> For example, I expected the following Gmsh commands
>>
>> step = 1.0;
>> For k In {0:100:step}
>> Printf("step = %f k = %f", step, k ) >> "test.txt";
>> step = step +1;
>> EndFor
>>
>>
>> would produce an output file called "test.txt" containing the
>> following data:
>>
>> step = 1.000000 k = 0.000000
>> step = 2.000000 k = 2.000000
>> step = 3.000000 k = 5.000000
>> .
>> .
>> .
>> and so on.
>>
>>
>> However, the previous commands result in the following output file:
>>
>> step = 1.000000 k = 0.000000
>> step = 2.000000 k = 1.000000
>> step = 3.000000 k = 2.000000
>> step = 4.000000 k = 3.000000
>> step = 5.000000 k = 4.000000
>> .
>> .
>> .
>> and so on.
>>
>>
>> It seems that incrementation of the "step" variable inside the loop
>> does not result in a variable update of "k", which keeps growing with
>> unit steps.
>>
>> I found this strange behaviour while trying to emulate a "While" loop
>> , which doesn't appear to be promptly available in Gmsh (by setting a
>> "step" variable bigger than the "for" loop's range, when some break
>> condition is satisfied ).
>>
>>
>> Am I missing something and these are in fact the expected behaviours,
>> or would this be a bug of some sort?
>
> Indeed, it's a limitation of our implementation: we don't reevaluate the
> arguments of the For after each iteration...
>
>
>> Is there a better way to work these issues around (emulating a "While"
>> loop) ?
>>
>
> Not that can I directly think of. I am adding this issue to our TODO list.
>
> Thanks for the feedback,
>
> Christophe
>
>>
>>
>> yours faithfully,
>> Lucas Blattner Martinho
>> **
>>
>>
>>
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> gmsh mailing list
>> gmsh at geuz.org
>> http://www.geuz.org/mailman/listinfo/gmsh
>
>


-- 
Prof. Christophe Geuzaine
University of Liege, Electrical Engineering and Computer Science
http://www.montefiore.ulg.ac.be/~geuzaine