<br><br><div class="gmail_quote">On Wed, Mar 12, 2008 at 10:08 AM, Paolo Bonzini <<a href="mailto:bonzini@gnu.org">bonzini@gnu.org</a>> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div class="Ih2E3d">Eliot Miranda wrote:<br>
> To be even more blunt I think the standards process would be damaged if<br>
> it is used as a way to introduce experimental extensions into the<br>
> language. I can imagine that attemting to standardise method pragmas<br>
> would be contentious enough given that theyre supported by only three<br>
> dialects to my knowledge, and then in different ways, even though<br>
> they're of proven use. But attempting to standardize something that<br>
> isn't in the base of any dialect is going to cause discord. IMO, the<br>
> time to standardise some new feature is when it is well understood, has<br>
> proven its utility in practice and has a functional supporting ecology<br>
> (tools etc). A-SIGNATURES doesn't meet any of these criteria.<br>
<br>
</div>IIUC, these "A-signatures" are just a way to hijack cons cells as a way<br>
to describe method signatures (and using chains of Associations as cons<br>
cells is often done in Workspace experiments -- I remember seeing it<br>
from you even while working on AOStA...). So I can imagine a "real"<br>
class MethodSignature to accept an "A-signature" as a shortcut, and this<br>
was the usage I envisioned in my previous message to Panu.</blockquote><div><br><br>OK, I misinterpreted. I i were looking for a shortcut I'd use literal arrays thus:<br><br>#(<font face="Courier New">
Message >> forwardTo: </font><font face="Courier New">Object </font><font face="Courier New">withArguments: Collection ^ Object)</font><br>
#(<font face="Courier New">
Message class >> selector: Object </font><font face="Courier New">^ Message)<br></font>
#(<font face="Courier New">
</font><font face="Courier New">Integer >> >> Integer ^ Integer)<br style="font-family: arial,sans-serif;"><br style="font-family: arial,sans-serif;"><span style="font-family: arial,sans-serif;">etc, so we still have infix selectors. So the receiver type is the token(s) up to the first #>> and the result type is the token(s) after #^.</span><br>
</font><br></div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">However, for the sake of "writing STEPs" I agree that we should use the<br>
protocol system used by the ANSI standard.</blockquote><div><br>Yep.<br> <br></div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><font color="#888888">Paolo<br>
</font><div><div></div><div class="Wj3C7c"><br>
_______________________________________________<br>
ANSI-Smalltalk mailing list<br>
<a href="mailto:ANSI-Smalltalk@lists.openskills.org">ANSI-Smalltalk@lists.openskills.org</a><br>
<a href="http://lists.openskills.org/cgi-bin/mailman/listinfo/ansi-smalltalk" target="_blank">http://lists.openskills.org/cgi-bin/mailman/listinfo/ansi-smalltalk</a><br>
</div></div></blockquote></div><br>