[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
- To: vanroggen%aitg.DEC@decwrl.dec.com
- Subject: Issue: DEFSTRUCT-REDEFINITION
- From: Jon L White <email@example.com>
- Date: Thu, 20 Oct 88 19:53:30 PDT
- Cc: firstname.lastname@example.org
- In-reply-to: vanroggen%aitg.DEC@decwrl.dec.com's message of Thu, 13 Oct 88 12:08:50 PDT <8810131908.AA13980@decwrl.dec.com>
re: I'd favor requiring redefining the same DEFSTRUCT as not being an error.
One way to tell that two DEFSTRUCTs are the same (although probably not
the way any implementation would really want to do it) it that two
calls to DEFSTRUCT are the same if they are EQUAL. It's hard to specify
something more general that might not get into problems with various
I'll echo your sentiments.
One can also have a class of errors called REDEFINITION, meaning that
it wasn't mechanically certifiable that the new definition is compatible
with the previous one. I wouldn't mind having the normal state for
this to be merely a warning. Lucid has a global variable named
*REDEFINITION-ACTION* which controls whether such redefinition cause
warnings, interactive queries (equivalent to continuable error), or
nothing at all.
Of course, compile-file might provider a "handler" for REDEFINITION.
-- JonL --