Am 02.03.21 um 21:32 schrieb Artur Tarassow:
Am 02.03.21 um 17:46 schrieb Riccardo (Jack) Lucchetti:
> On Tue, 2 Mar 2021, atecon wrote:
>
>>
>>
>> Am 02.03.2021 15:11 schrieb Sven Schreiber:
>>> Am 26.02.2021 um 10:51 schrieb Sven Schreiber:
>>>> Am 19.02.2021 um 13:52 schrieb Allin Cottrell:
>>>>> On Fri, 19 Feb 2021, Riccardo (Jack) Lucchetti wrote:
>>>>> I can't think of an easier way of doing this, and your function
>>>>> seems to
>>>>> me a good candidate for "extra". One possible modification
would be:
>>>>>
>>>>> function matrix unstack(series x, series names[null], \
>>>>> strings S[null])
>>>>>
>>>>> where the caller needs to supply one of @names or @S, allowing
>>>>> for the
>>>>> possibility that there's no handy series holding group names and
you
>>>>> therefore want to submit an array of group names directly. Or,
>>>>> drop the
>>>>> @S argument but be ready to construct automatic names (e.g.
>>>>> "unit%d") in
>>>>> case @names is not given.
>>>>>
>>>>
>>>> Jack, would you take care of pushing this into 'extra', or is
further
>>>> discussion needed?
>>>>
>>> Please, what's the status here?
>>
>> Hi,
>>
>> I asked Jack to think about the possibility to make this a (I think)
>> natural extension of my user-contributed "stack_data" package. But
>> no reply, yet ;-)
>
> Uhm, I think I'd rather have it inside "extra"; but I don't have a
> strong opinion on this, really.
I started to add it to stack_data. Let me show you the prototype
tomorrow or so, Jack. Then you can decide ;-)
Hi Jack,
here is the implementation of unstack():
https://github.com/atecon/stack_data/blob/master/src/stack_data.inp
Things are also tested:
https://github.com/atecon/stack_data/blob/126b04104f412bed4e2cfb8c8c72474...
Artur