Re: Method Dispatch and stuff.
Joe Knapka <[email protected]> 29 Mar 2004 08:56:56 -0700
| Newsgroups | gmane.comp.lang.prothon.devel |
|---|---|
| Message-ID | <[email protected]> |
Kyle Lahnakoski <[email protected]> writes: > What type of (dynamic) dispatch is used by Prothon? For example, does > prototype delegation stand in for inheritance when determining what > method signature best matches a given call signature. Is there > anything interesting to say about the conflict resolution (should more > than one method be a candidate for a given call)? If I understand correctly: There are no signatures, only names (just like Python, and *unlike* Java, C++, etc). Thus there is no possibility of a conflict. You search the proto heirarchy for an attribute with the name being invoked, and try to call it. If that fails, you're done, and the caller sees an exception. -- Joe > David MacQuigg wrote: > > >> Recently on the Python mailing list, I have been pushing the idea > >> that we need a special symbol ( @ $ or whatever ) as a > >> general-purpose "extension syntax". > > Please NO! :) > > I prefer a mandatory statement indicating the particular compiler > (language version). This statement could be a simple keyword-value > pair (eg VERSION 1.1), and could be put anywhere (but usually at the > top of a source file). Of course the VERSION keyword would obey > lexical scope rules. The advantages are clear: > * Nearly complete freedom from legacy code > * Shorter keywords (remove the magic symbol) > > > Thanks. > > > -- > ---------------------------------------------------------------------- > Kyle Lahnakoski [email protected] > (416) 892-7784 Arcavia Software Ltd > > > > _______________________________________________ > Prothon-dev mailing list > [email protected] > http://lists.prothon.org/mailman/listinfo/prothon-dev > > -- Resist the feed. -- If you really want to get my attention, send mail to jknapka .at. kneuro .dot. net.