[Haskell] Proposal: Relative Module Imports

[Simon Marlow]

On 03 May 2005 22:13, S. Alexander Jacobson wrote:

> Problem: We need a way to simplify module imports.
> 
> Problem details:
> 
> * Hierarchical module names are getting really long (including a
> functional area, a package name, and a module name).
> 
> * People typically import multiple modules from areas close to each
> other in the hierarchical module namespace (especially in the case of
> intra-package imports).
> 
> * Long module names are required even for non-exposed modules because
> a program may contain only one module with a given name (regardless of
> its visibility).
> 
> Idea: Allow module relative imports in a manner that does not break
> any existing code.
> 
> Proposal:
> 
> * Use preceding dots to indicate that module name is relative
> * Use from keyword to specify a different relative base.
> 
> Example:
> 
>   Dot relative syntax             Translation
>   -------------------             -----------
>   module Text.Space.Foo.M where   module Text.Space.Foo.M where
>   import .M2                      import Text.Space.Foo.M2 as M2
>   import ..Bar.Baz                import Text.Space.Bar.Baz as Bar.Baz
>   import Data.Set                 import Data.Set
>   from ...HaXML.XML
>     import .Types                 import Text.HaXML.XML.Types as Types
>     import .Escape                import Text.HaXML.XML.Escape as
>     Escape import .Pretty                import Text.HaXML.XML.Pretty
> as Pretty 
> 
> I believe that the proposed syntax is much more concise and readable
> than the current equivalent.

I was sure something like this had been suggested before, and in fact
several similar schemes have.  Here's a couple of starters:

http://www.haskell.org/pipermail/libraries/2001-February/000268.html
http://www.haskell.org/pipermail/libraries/2001-March/000322.html

Why haven't we ever implemented anything like this?  A good question,
with no really good answer.  I think it's a combination of 

  (a) the current situation isn't *really* hurting that much
  (b) the current situation is *really* easy to describe and implement
  (c) none of the proposed solutions are obviously the right thing
   
(e.g. the '.' prefixes look a little obscure, IMO).

and from my perspective:

  (d) I hoped that something like grafting would provide a more
      general solution.

Oh, and there was a recent proposal on the libraries list to allow
exporting of qualified names, which solves a similar/overlapping
problem:

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

FWIW, I think this proposal ended in a reasonable result.  It just needs
someone to implement it...

Cheers,
	Simon