BUG DESCRIPTION:
----------------
I notice that:
typeof(1:$) is (in Scilab 6) equal to "implicitlist", same for typeof(:).
[in 5.5.2 the types were "size implicit" and "constant" respectively, which was not very logical].
There may be some backward compatibility issues, but that's OK if the changes are clearly described
(if the changes make Scilab better designed).
1) I still wonder about the "$" case:
typeof($) => "polynomial" (in Scilab 6 and previous versions)
From a user's point of view that does not make much sense:
what is the difference between "$" and "$:$" ?
Shouldn't the type be changed as for "1:$" and ":" ?
[it would make code in overloaded functions less obscure]
2) Also (in Scilab 6) :
-->typeof(:)
ans =
implicitlist
-->disp(:)
1:1:$
-->:
[Scilab6] :
[Scilab6] ^^
[Scilab6] prompt : 1.1 - 1.2 : syntax error, unexpected :
If ":" has a type and can be displayed, what cannot be evaluated directly ?
That's not very logical.
What do you think about points 1 and 2 ?