[mercury-users] bug in mmc --aditi

[Fergus Henderson]

On 05-Aug-2001, Holger Krug <hkrug at rationalizer.com> wrote:
> > ...
> 
> Thank you for the info !
> 
> > If you want sources which exactly match a given binary snapshot,
> > then you'll need to use CVS.
> 
> Can you give me a cvs read account ? Or did I fail to notice an
> anonymous cvs account ?

We provide read-only anonymous CVS access.  See
<http://www.cs.mu.oz.au/research/mercury/information/developers/remote_cvs.html>.

> It would be even simpler for the reviewers,
> because I could send in patches based on the current cvs status and with
> the correct cvs repository info inside the patch.

Yes, this is strongly encouraged ;-)

> You work with several different cvs branches, I suppose ? On which to base
> my patches (concerning lex, moose, and store.m) ?

The main branch.

For more information about how to contribute to the Mercury project, see 
<http://www.cs.mu.oz.au/research/mercury/information/developer.html>.

> The new typeclass-based version of store.m I want to call store_new.m,
> to make clear by its name, that it is a replacement of store.m.
>
> When I send the patch for store_new.m, I have to mark store.m deprecated. There
> is a pragma `obsolete' on a per predicate base. Is their a similar pragma for
> whole modules ?

I'm afraid not.  The best you can do currently is to mark all of the functions
and predicates in the old module as deprecated.

> Should I change `store' to `store_new' all-over the stdlib and add store_new.m
> to the Makefile ?

It would be nicer if we had a better solution to these kind of source-level
backwards compatibility issues.  It would be nice, for example,
we had a compiler option that you could use when compiling modules that
import `store' that would make make them use `store_v1' instead.
Then you could just rename the old version of store.m as store_v1.m.
New code could continue to use the natural name `store'.

However, in this particular case, I have an idea for how you can make
the changes to store.m in a way that preserves backwards compatibility.
I will explain on the mercury-developers list.

-- 
Fergus Henderson <fjh at cs.mu.oz.au>  |  "I have always known that the pursuit
The University of Melbourne         |  of excellence is a lethal habit"
WWW: <http://www.cs.mu.oz.au/~fjh>  |     -- the last words of T. S. Garp.
--------------------------------------------------------------------------
mercury-users mailing list
post:  mercury-users at cs.mu.oz.au
administrative address: owner-mercury-users at cs.mu.oz.au
unsubscribe: Address: mercury-users-request at cs.mu.oz.au Message: unsubscribe
subscribe:   Address: mercury-users-request at cs.mu.oz.au Message: subscribe
--------------------------------------------------------------------------