On Mon, Jul 11, 2011 at 11:48 PM, Ivan Lazar Miljenovic wrote:

...

A small standalone program to manage the import block of a haskell program. It will try to add import lines for qualified names with no corresponding import, remove unused import lines, and sort the import block according to some heuristic you can define. This only works for qualified imports! Unqualified imports are left untouched.

It's most convenient if bound to an editor key.

This sounds interesting; it would be helpful if you could provide an example if you could provide an example in the README or description of what exactly this program does however.

Sure, well... it's like the paragraph above :) As an example, I just hack away in haskell normally. When I've edited a chunk, and have a feeling a changed the dependencies, I hit ',a' in vim, and it will remove imports that are no longer used, and add imports that are now needed. As a bonus it'll sort and format the import list. That way I don't have to figure out if there are new modules needed, go to the top of the file, sift past the module haddock and find the right spot in the potentially large import list (and maybe I can't just ! out to 'sort' for that if some are qualified and some are unqualified), and type all that repetitive 'import qualified Blah.Blah.Blah as Blah' stuff. And I don't have to wait for ghc to warn me about redundant imports and then go delete them one by one. In addition to being generally convenient, I've found I'm more willing to use tiny helper functions like 'Maybe.fromMaybe' instead of just 'maybe x id' since the hassle of possibly adding the new import has been eliminated. It's a standalone program so it doesn't have to be vim, but that's what I use, so that's what I include binding glue for. I've also got an unrelated simple thing that keeps the currently edited haskell module in a file so I can type :L in ghci to load the module currently being edited. Very convenient, especially when modules start getting nested.

I meant more in the "Before: ... After: ..." sense ;-) (i.e. visually being able to tell what your program does, rather than just a description).

Oh, ok, I guess I can do that too: Before: module M where import ZZ.Out.Of.Order import qualified No.Longer.Needed as Needed -- x = Needed.y -- just removed this y = New.something -- just added this After: module M where import qualified Heavily.Nested.New as New import ZZ.Out.Of.Order -- x = Needed.y -- just removed this y = New.something -- just added this

On Wed, Jul 13, 2011 at 12:08 PM, Henning Thielemann wrote:

I also spend some time on managing imports. I'll try this program. However, usually I have a mix of qualified and unqualified imports, since some libraries are designed for unqualified imports.

You can still use unqualified imports, it just won't manage those for you. Automatically managing unqualified imports is solvable but harder.

On Wed, Jul 13, 2011 at 3:05 PM, Markus Läll wrote:

One thing that to look out for though is that it first checks for modules from the current directory down -- so running it in somewhere you happen to have ghc sources too, will have all that come with it imported from there (and lines like "import path.to.ghc.DataMaybe (fromJust)" added). But this probably happens only when testing the program in the root of your home directory...

Yeah, currently the most annoying thing is when it guesses wrong. It's set up to work with my project, but you can tweak the search heuristic easily enough by editing findLocalModule. I'll have it skip non-capitalized directories, but a better solution is to not edit modules in your home directory :)

(The reason I'm not using it regularly is that I haven't figured out a good way to use it, when developing with Notepad++ over sftp in another environment)

Well, if anyone makes bindings for other editors, feel free to send a patch.