Re: package mounting

I like your proposal and I think that some form of it should be implemented. Here is the section of my original post where I talk about it:

One further thing, there have been proposals to simplify the importing of collections of modules from a certain point in the namespace, etc. I hope it is realized that they are independent from my proposal. They would not be very useful in implementing my proposal, at least I think any such solution would be far from optimal; and vice-versa. Modules and packages are quite distinct constructs, modules are needed for namespace partitioning and packages are needed to delineate administrative boundaries and sources of change. Both are necessary and both deserve special consideration in the ongoing design of Haskell.

I still feel this way. Obviously there is some overlap in functionality but unless you want to either bind what's normally under Graphics.UI.Gtk at the namespace root everywhere in the project, or mount the library at some sort of short prefix like Gtk and use Gtk.Button.label everywhere - and neither of these is really suitable - one can't use package mounting to implement your proposed feature. And as with any other feature which exists within the language, your proposal can't distinguish between modules of the same name coming from different packages, so it couldn't be used to build a program's module namespace by mounting individual package namespaces at specified locations (and allowing these package namespaces to have modules of the same name defined), which is what I'm trying to do. It's like the difference between filesystem mounting and a PATH environment variable. Hope this makes sense. Frederik On Mon, Jun 27, 2005 at 12:20:04PM +0100, Malcolm Wallace wrote:

"Simon Marlow" writes:

...

, Frederik Eaton wrote:

...

It looks like there's been a bit of recent discussion regarding module and package namespaces.

This idea has been raised before, but it was a while back, and we called it "grafting".

I concur with Simon that the mounting proposal has similarities to the grafting proposal. I also concur that opinions may have changed since the previous discussion, so it is worthwhile bringing it up again.

I also wonder whether (or to what extent) mounting/grafting could address (or conflict with) the question of qualified name exports expressed in this thread:

http://www.haskell.org//pipermail/libraries/2004-December/002800.html http://www.haskell.org//pipermail/libraries/2005-March/003390.html

Regards, Malcolm _______________________________________________ Libraries mailing list Libraries@haskell.org http://www.haskell.org/mailman/listinfo/libraries

-- http://ofb.net/~frederik/