[m-dev.] insts for existential data

[Julien Fischer]

On Tue, 14 Nov 2006, Mark Brown wrote:

> When specifying the inst for an existentially typed data constructor, are
> we expected to add a `ground' argument for the automatically inserted
> type_info?  I would have thought not, but leaving it out leads to mode
> errors.

I agree; requiring users to add an inst for the type_info unncessarily
exposes implementation details.  (So the above mode errors are compiler
bugs.)

Julien.
--------------------------------------------------------------------------
mercury-developers mailing list
Post messages to:       mercury-developers at csse.unimelb.edu.au
Administrative Queries: owner-mercury-developers at csse.unimelb.edu.au
Subscriptions:          mercury-developers-request at csse.unimelb.edu.au
--------------------------------------------------------------------------