Hi Samuel,
Le 10/08/2016 à 20:02, Samuel Gougeon a écrit :
Hello Pierre-Aimé,
Le 10/08/2016 17:56, Pierre-Aimé Agnel a écrit :
Hello,
It was changed at resolution of
http://bugzilla.scilab.org/show_bug.cgi?id=14316
I am not sure though whether enforcing the 'vector ^ scalar' error
should be reintroduced. My 2-cents: there are many operations that
have the same behaviour that their dot-operation counterpart when
used with a scalar.
I think the operation should remain available, but I can also enforce
the obsolescence in the release for 6.0.0.
Thanks for your answer.
I understand that the reversion of the vector^scalar change was done
when fixing #14316,
but as far as i understand, things are not connected. This reversion
was not mandatory to fix #14346.
IMO, what was planed for Scilab 6 was great.
I know that this syntax vector^scalar instead of vector.^scalar was
very widespread. But instead of parsing vector^scalar as
vector.^scalar as in Scilab <6 or always yielding an error, why not
making Scilab 6 testing whether the overload %s_p_s() (or
other%@_p_@() overload according to operands types) exists. Then,
users wanting to work with the old equivocal behavior could define
deff("r=%s_p_s(a,b)","r=a.^b") in their library (or startup file)
instead of changing all their code. Or/and include their own warning
in their overload whether they wish to actually upgrade their code.
And that's it.
In my humble opinion, this is a great proposal that can benefit
everyone, we just have to check the already existing overloads to see if
there are any interferences.
Best regards
Samuel
_______________________________________________
dev mailing list
[email protected]
http://lists.scilab.org/mailman/listinfo/dev
--
Pierre-Aimé Agnel
R&D Projects Manager
Phone: +33.1.80.77.04.67
Mobile: +33.6.82.49.35.23
-----------------------------------------------------------
Scilab Enterprises
143bis rue Yves Le Coz - 78000 Versailles, France
Phone: +33.1.80.77.04.60
http://www.scilab-enterprises.com
_______________________________________________
dev mailing list
[email protected]
http://lists.scilab.org/mailman/listinfo/dev
