On Thu, Apr 05, 2001 at 04:18:14PM -0500, Steven Hazel wrote: > The problem with thinking of keys in terms of function calls is that > in order to spec it out fully, we'd have to invent a type system. The > return types on AGL's "functions" weren't specified in his email, but > they did exist -- the DBR() function, for example, has a return type > that isn't the same as that of either of the parameters taken by the > CHK() function.
Well, I certainly hinted at the return values and I certainly intended a type system. > A more limited approach would save us all the trouble of embedding > type checkers in our freenet URI parsers. Yes, it would. And as I've said before the Rambling system is flexable in the extreme. I really don't think implimenting the type sytem would be difficult at all. If you want to start putting DBRs in the keys that's a major decision and we need to have some kind of flexable system. AGL -- Never attribute to malice that which is adequately explained by stupidity. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 240 bytes Desc: not available URL: <https://emu.freenetproject.org/pipermail/devl/attachments/20010406/5c3215e9/attachment.pgp>
