Path: utzoo!attcan!uunet!cs.utexas.edu!usc!brutus.cs.uiuc.edu!apple!oracle!news From: rdavis@oracle.com (Richard Davis) Newsgroups: comp.object Subject: Re: Shippable C++ Objects (RFC) Message-ID: Date: 18 Nov 89 02:08:10 GMT References: <31.UUL1.3#5109@pantor.UUCP> Sender: news@oracle.com Organization: Oracle Corperation, Belmont, CA Lines: 21 In-reply-to: richard@pantor.UUCP's message of 16 Nov 89 13:43:31 GMT > richard@pantor.UUCP (Richard Sargent) says: > > "The object can reconstruct itself from the byte stream" requires > that the object already exist. It still seems to me that you *have* > to have a case statement somewhere which knows about all (interesting) > object classes. The cases create each new type of object, probably using > a class constructor which has a ByteStreamRepresentation argument. > The created object is then added into the receiving program's > organization. How to do this from a general purpose class library > without encoding knowledge of the application is another (lesser) > problem. Actually, it presents no more of a problem (and no less of one :-) than the bootstrapping problem of loading a disk operating system from disk. The ByteStreamRepresentation needs to contain a meta-represen- tation (yuck) of the structure of the object to be created, which can be parsed by the constructor on the receiving end. Richard Davis Internet: rdavis@oracle.com