Expected Common Misunderstanding (was: bug: run/3 bug)

Mark S. Miller markm@caplet.com
Mon, 07 Jun 1999 19:20:38 -0700


At 08:55 AM 6/7/99 , Ben Laurie wrote:
>You could simply generate a warning when in interactive mode, i.e.:
>
>build3(a,b,c) has splatted existing build3(a,b).
>
>Or is that too simpleminded?

At 08:36 AM 6/7/99 , Dean Tribble wrote:
>The super-clever idea is just more diagnostics.  In the example, the
>compiler could have issued a warning that the build3 function was being
>redefined (and let it proceed).  

You're both right, and Thanks!  I'm glad there's an adequate answer this 
simple!  I'll do this.


>It could also have complained that the
>build3/3 call wasn't defined (again, while accepting the definition).  This
>latter analysis is obviously harder!  

Yup.  Go down that road, and before you know it you're trying to do static 
type inference.  Which would be really cool, but isn't necessary for any of 
my goal for the next few years.  However, this is a Bazaar-style open source 
project.  Anyone out there want to explore in this direction?