<html>
<head>
<meta content="text/html; charset=windows-1252"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<div class="moz-cite-prefix">Le 12/02/2018 à 21:30, Samuel Gougeon a
écrit :<br>
</div>
<blockquote cite="mid:ac718b95-9081-0ffd-097c-4436e9c4332a@free.fr"
type="cite">
<meta http-equiv="content-type" content="text/html;
charset=windows-1252">
<p>Dear ESI devs,</p>
<p>In Scilab < 6, typeof(1:$) was "<code class="literal">size
implicit"</code> </p>
<div class="refsection">In Scilab 6.0, typeof(1:$) has become
"implicitlist"<font face="Arial"><br>
<br>
</font>Why not.<br>
The issue is that the overloading code is still <tt>"ip"</tt>
standing for <tt>"implicit polynomial"</tt><font face="Arial">,</font>
not<font face="Arial"> </font><tt>"il"</tt><font face="Arial">.<br>
</font>And in practical, this object is actually made of
polynomial(s)<br>
<br>
So, in one hand, for the same object, we refer to lists. On the
other one, we refer to polynomials.<br>
I don't know what was the motivation in renaming "<code
class="literal">size implicit"</code> into <tt>"implicitlist"</tt><font
face="Arial">.</font><br>
If it was to remove an inconsistency somewhere, it apparently
adds another one elsewhere.<br>
<font face="Arial"><br>
</font>So: why not<tt> </tt><tt>typeof(1:$,"overlaod")=="il" </tt><font
face="Arial">?</font></div>
</blockquote>
<br>
Actually, i think that "implicitlist" is right.<br>
"Size implicit" emphasized one usage (among other possible ones) :
indexation.<br>
"implicit polynomial" points to the components of the object, more
than the object itself.<font face="Arial"><br>
<br>
</font>So this renaming is rather welcome and relevant. But now it
would be even better to get a consistent overloading code.<br>
<br>
Samuel<br>
<br>
</body>
</html>