[r6rs-discuss] [Formal] Trivial Enhancement of macros in v5.91:
aghuloum at cs.indiana.edu
Thu Nov 30 09:20:22 EST 2006
On Nov 30, 2006, at 8:01 AM, William D Clinger wrote:
> With the shared binding, invoke-once semantics,
> non-generative record types work as programmers
> expect, but the draft R6RS doesn't guarantee that
What do programmers expect from non-generative records?
Putting a programmer hat, I expect records instantiated
from non-generative types to be the same every time (when
a library is compiled, visited, invoked, and executed in
a script). Is this consistent with your assumptions?
More information about the r6rs-discuss