[Scilab-Dev] varn([]) changed

[Antoine ELIAS]

I'm not sure that empty rlist is properly managed in Scilab

try
a = rlist() //without ;

and overload "%r_varn" tries to access to internal fields that does not 
exist.
I think the trouble comes from rlist function.

Antoine
Le 15/01/2019 à 10:29, Samuel Gougeon a écrit :
> Hello Antoine,
>
> Le 15/01/2019 à 09:53, Antoine ELIAS a écrit :
>> Hello Samuel,
>>
>> No, it is not intentional, I will take a look to that today.
>
> IMO, this is not necessarilly a bad change. Please also consider that 
> remark <http://bugzilla.scilab.org/show_bug.cgi?id=11077#c8> about 
> applicability of varn() to empty rlist().
>
> Thanks
> Samuel
>
>>
>> Antoine.
>> Le 14/01/2019 à 23:22, Samuel Gougeon a écrit :
>>> Hello,
>>>
>>> I was noting that varn([]) changed from Scilab 5 to Scilab 6:
>>>
>>> 5.5.2:
>>> -->varn([])
>>>          !--error 246
>>> Function not defined for given argument type(s),
>>>   check arguments or define function %s_varn for overloading.
>>>
>>> 6.0.1:
>>> --> varn([])
>>>  ans  =
>>>     []
>>>
>>> This is hard-coded, not in any %s_varn().
>>>
>>> This change is not yet documented. I would like to have confirmation 
>>> that it is intentional.
>>>
>>> Thanks
>>> Samuel 
>
>
> _______________________________________________
> dev mailing list
> dev at lists.scilab.org
> http://lists.scilab.org/mailman/listinfo/dev

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.scilab.org/pipermail/dev/attachments/20190115/eefd3ed0/attachment.htm>