FW: [commit: ghc] tc-arrows: Rearrange the typechecking of arrows, especially arrow "forms" (49badc9)

Ross
I've pushed a new branch, tc-arrows, for the new typechecking of arrows, based on the discussion we had last week.
It all compiles, but won't work because I have done nothing to the desugaring.
It uses our new representation with (env, nested-tuple-with-unit-at-end)
rather than (nested-tuple-with-env-at-the-end), so desugaring needs significant attention.
Would you like to work on the desugarer? I'm out of my depth, but happy to advise.
Thanks
Simon
| -----Original Message-----
| From: ghc-commits-bounces@haskell.org [mailto:ghc-commits-
| bounces@haskell.org] On Behalf Of Simon Peyton Jones
| Sent: 14 February 2013 16:08
| To: ghc-commits@haskell.org
| Subject: [commit: ghc] tc-arrows: Rearrange the typechecking of arrows,
| especially arrow "forms" (49badc9)
|
| Repository : ssh://darcs.haskell.org//srv/darcs/ghc
|
| On branch : tc-arrows
|
| http://hackage.haskell.org/trac/ghc/changeset/49badc91c52fc1f65a3ff5bfb3
| edaecbd617de61
|
| >---------------------------------------------------------------
|
| commit 49badc91c52fc1f65a3ff5bfb3edaecbd617de61
| Author: Simon Peyton Jones

On Thu, Feb 14, 2013 at 05:40:51PM +0000, Simon Peyton-Jones wrote:
I've pushed a new branch, tc-arrows, for the new typechecking of arrows, based on the discussion we had last week.
It all compiles, but won't work because I have done nothing to the desugaring.
It uses our new representation with (env, nested-tuple-with-unit-at-end) rather than (nested-tuple-with-env-at-the-end), so desugaring needs significant attention.
I think it's done. All but 4 of the tests pass: - I have new versions of arrowform1 and T5283 that pass. - The type error message for T5380 has changed. - The gimmick used by arrowapply4 is incompatible with the new representation, so this test should be dropped.

Sounds great. Will you send me the new versions of arrowform1 and T2583? Thanks Simon | -----Original Message----- | From: Ross Paterson [mailto:ross@soi.city.ac.uk] | Sent: 19 February 2013 13:41 | To: Simon Peyton-Jones | Cc: ghc-devs@haskell.org | Subject: Re: FW: [commit: ghc] tc-arrows: Rearrange the typechecking of arrows, | especially arrow "forms" (49badc9) | | On Thu, Feb 14, 2013 at 05:40:51PM +0000, Simon Peyton-Jones wrote: | > I've pushed a new branch, tc-arrows, for the new typechecking of arrows, based | on the discussion we had last week. | > | > It all compiles, but won't work because I have done nothing to the desugaring. | > | > It uses our new representation with (env, nested-tuple-with-unit-at-end) | > rather than (nested-tuple-with-env-at-the-end), so desugaring needs significant | attention. | | I think it's done. All but 4 of the tests pass: | - I have new versions of arrowform1 and T5283 that pass. | - The type error message for T5380 has changed. | - The gimmick used by arrowapply4 is incompatible with the new representation, | so this test should be dropped.

Ross Excellent. We have moved forward nicely on the arrows front. I have pushed the refactoring to typechecking arrows that we did together. Remaining: * You missed one test failure: ghci/scripts T5045 [bad stderr] (ghci) I'm sure that just looking at the source code will tell you why it no longer typechecks, and the fix will be simple. Could you spare a moment to fix this soon? At the moment it'll make validate fail. I could open a ticket and make it expect_broken, but it'd be less of a hassle to fix it immediately. 5 min job. The remaining open tickets I know about are: * #5267 * #5777 * #5333 * #344 Might you look at them? Thanks Simon | -----Original Message----- | From: Ross Paterson [mailto:ross@soi.city.ac.uk] | Sent: 19 February 2013 13:41 | To: Simon Peyton-Jones | Cc: ghc-devs@haskell.org | Subject: Re: FW: [commit: ghc] tc-arrows: Rearrange the typechecking of | arrows, especially arrow "forms" (49badc9) | | On Thu, Feb 14, 2013 at 05:40:51PM +0000, Simon Peyton-Jones wrote: | > I've pushed a new branch, tc-arrows, for the new typechecking of | arrows, based on the discussion we had last week. | > | > It all compiles, but won't work because I have done nothing to the | desugaring. | > | > It uses our new representation with (env, | > nested-tuple-with-unit-at-end) rather than (nested-tuple-with-env-at- | the-end), so desugaring needs significant attention. | | I think it's done. All but 4 of the tests pass: | - I have new versions of arrowform1 and T5283 that pass. | - The type error message for T5380 has changed. | - The gimmick used by arrowapply4 is incompatible with the new | representation, | so this test should be dropped.
participants (2)
-
Ross Paterson
-
Simon Peyton-Jones