Ticket #1127: proposal-1127-1.patch

File proposal-1127-1.patch, 1.6 KB (added by igloo, 8 years ago)
Line 
1-----BEGIN PGP SIGNED MESSAGE-----
2Hash: SHA1
3
4
5New patches:
6
7[Make StateT lazy; trac proposal #1127
8Ian Lynagh <[email protected]>**20070130165252] {
9hunk ./Control/Monad/State.hs 195
10- -             (a, s') <- runStateT m s
11+               ~(a, s') <- runStateT m s
12}
13
14Context:
15
16[update broken URL as noticed by [email protected]
17Simon Marlow <[email protected]>**20070118094739]
18[Added Haddock documentation. Converted the module documentation to Haddock format. Per Jeff Newbern's permission included parts his tutorial "All About Monads" http://www.nomaware.com/monads/.
19Andriy Palamarchuk <[email protected]>**20061218165621]
20[add boilerplate Setup.hs
21Ross Paterson <[email protected]>**20060928231525]
22[use Control.Monad.Instances from base
23Ross Paterson <[email protected]>**20060410112533]
24[Add -fallow-undecidable-instances to some Control.Monad modules
25[email protected]**20060209121334
26 
27 I have recently tightened up GHC's implementation of the coverage
28 condition.  As a result some of the Control.Monad modules are rejected.
29 
30 Example:
31   class (Monad m) => MonadReader r m | m -> r where
32   instance (Monoid w, MonadReader r m) => MonadReader r (WriterT w m)
33 Here, fv(Writer w m) is not a superset of fv(r).
34 
35 The flag allows it.  I wonder if it's possible to use these modules
36 to send the type cheker into a loop.
37 
38 
39]
40[TAG Initial conversion from CVS complete
41John Goerzen <[email protected]>**20060112154134]
42Patch bundle hash:
438de1c0d7cdb1faa09c0783ba3db5ae9abc166a04
44-----BEGIN PGP SIGNATURE-----
45Version: GnuPG v1.4.1 (GNU/Linux)
46
47iD8DBQFFv3g60y4LitoVXIcRAp5ZAKDsHUNLp4+rnmYimTtmsJZ8UOtssQCggkUi
487R/LT8bO/C483TudX3FtSNE=
49=zpqL
50-----END PGP SIGNATURE-----