Re: [Quantlib-users] Cloning an overnightIndex

[David D. <nh...@gm...>]

Thanks Luigi. Issue added on GitHub.

On Tue, 21 Jul 2020 at 09:36, Luigi Ballabio <lui...@gm...>
wrote:

> Hello,
>     it's kind of forced on us by the need of virtual functions to have the
> same return type.  In C++ one can easily downcast the pointer to the
> correct class again, but that doesn't carry well to Python.  We should
> override the method in the SWIG interface so it returns the more derived
> type.  May you open an issue on GitHub?
>
> Thanks,
>     Luigi
>
>
> On Tue, Jul 21, 2020 at 1:14 AM David Duarte <nh...@gm...> wrote:
>
>> Hi all,
>>
>> In python, cloning an Ibor index seems to be a good way to assign a
>> YieldTermStructure after it is created.
>>
>> For example: index.clone(yts)
>>
>> However this doesn't work with an overnight index because after cloning,
>> the index type changes to IborIndex.
>>
>> For example,  creating and index with ql.FedFunds() shows :
>>
>> <QuantLib.QuantLib.FedFunds; proxy of <Swig Object of type 'boost::shared_ptr< FedFunds > *' at 0x7f0706751a80> >
>>
>>
>> But if I clone it, with ql.FedFunds().clone(yts), the instance changes to:
>>
>>
>> <QuantLib.QuantLib.IborIndex; proxy of <Swig Object of type 'boost::shared_ptr< IborIndex > *' at 0x7f0706751de0> >
>>
>>
>> Is this expected?
>>
>>
>> _______________________________________________
>> QuantLib-users mailing list
>> Qua...@li...
>> https://lists.sourceforge.net/lists/listinfo/quantlib-users
>>
>