Path: news.daimi.aau.dk!not-for-mail From: John Cowan Newsgroups: comp.lang.beta Subject: Re: Loki Paper 1: Overview Date: Fri, 20 Mar 1998 08:43:07 +0100 (MET) Organization: DAIMI, Computer Science Dept. at Aarhus University Lines: 54 Approved: mailtonews@daimi.aau.dk Distribution: world Message-ID: <199803200743.IAA24530@noatun.mjolner.dk> Reply-To: John Cowan NNTP-Posting-Host: daimi.daimi.aau.dk Xref: news.daimi.aau.dk comp.lang.beta:11474 Erik Ernst writes: > You intend to translate to Java or to JVM byte-code? Java itself. I think there is nothing gained by the use of bytecode, except compile-time efficiency. > Anyway, I think > that the type verification phase will demand equivalent casts in byte > code as the ones needed for a BETA-source to (correct) Java-source > translation. Yes. > Do you plan to use fixed classes in place of virtual classes and then > insert downcasts wherever the virtual class (according to the BETA > static analysis) is known to be more specialized? That's like: No, I plan to map virtual patterns to methods that return Class objects. In that way, the virtuality works correctly: you can instantiate with aClass.virtualPattern().newInstance() and call methods using Java reflection. This will be explained in a later paper. > You cannot annotate the occurrence of `v' in the exit list of `p' with > a cast, because the statically known type of the exited value differs > depending on _the static knowledge about the enclosing p/q/... object > at the point of the evaluation_. I will consider your examples in detail at a later time. I am saving all mail received. > evaluation (semi-ignoring the multiple-return-value problem): Multiple values are handled by the class Tuple_, which is part of the implementation machinery: it essentially abstracts a (Java) array of Object (references). > But the entire fragment language could be ignored to get started, > because no BETA program "essentially" needs it. Just transform the > program using cut-n-paste until it's all in one file and all SLOTs are > gone. That is my intention. > (Well, there are some name clash issues, too, but it almost > works like that :-) I need to fully understand just which names in BODY fragments are hidden. When I understand that, name clash removal will be very straightforward. -- John Cowan cowan@ccil.org e'osai ko sarji la lojban.