Re: Any documentation of changes to libinterp API?

[Orion Poplawski]

On 01/01/2014 11:54 PM, John W. Eaton wrote:
> On 01/01/2014 10:55 AM, Orion Poplawski wrote:
>> On 12/31/2013 10:34 PM, Carnë Draug wrote:
>>> On 1 Jan 2014 03:37, "Orion Poplawski" <address@hidden> wrote:
>>>>
>>>> Is there any documentation of changes to the libinterp C++ API?
>>>> [...]
>>>> So it appears that is_real_nd_array() is gone.  What replaces it!
>>>
>>> Doxygen documentation has already been updated [1].
>>>
>>> Carnë
>>>
>>> [1] http://octave.sourceforge.net/doxygen/html/index.html
>>>
>>
>> Thanks.  But while that documents the current state of things, it
>> doesn't appear to have a "changes" section that would help with
>> migrating code.
>>
>> I guess we'll actually need to read the code and figure out what it is
>> trying to do :).
> 
> Looking at the 3.6.4 sources, I'm not sure that is_real_nd_array was ever
> doing anything but returning false.  I only see the function in ov.h that
> forwards to rep, and then generic definition of it in ov-base.h that always
> returns false.  There don't seem to be any other overloads for it.
> 
> I guess I would use is_real_type.  You should be able to extract an NDArray
> object from any real-valued object (if not, then that is a bug).  In most
> cases, I'd also recommend that instead of checking the type first, that you
> just try to extract an NDArray object and then check error_state to see
> whether it succeeded.
> 
> jwe
> 
> 

Thanks, I've passed that on to the swig folks.



-- 
Orion Poplawski
Technical Manager                     303-415-9701 x222
NWRA, Boulder/CoRA Office             FAX: 303-415-9702
3380 Mitchell Lane                       address@hidden
Boulder, CO 80301                   http://www.nwra.com