[m-dev.] non-unique I/O globals field

[Ralph Becket]

Julien Fischer, Wednesday,  7 February 2007:
> 
> Are there are any objections to making the I/O globals field non-unique?
> It doesn't really work for it to be unique (see various XXX comments
> through the library and compiler) and having it unique means that
> programmers more often than not end up calling unsafe_promise_unique
> everytime they wish to use it.
> 
> The change shouldn't be too distruptive (hopefully a lot of uses
> of I/O globals has been migrated to mutables now anyway.)

Go for it.
--------------------------------------------------------------------------
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
--------------------------------------------------------------------------