search

jgm / cheapskate

Experimental markdown processor in Haskell
BSD 3-Clause "New" or "Revised" License
107 stars 18 forks source link

GHC 7.10 fixes #9

Closed bgamari closed 9 years ago

bgamari commented 9 years ago

It appears that this is already fixed in master but has yet to be pushed to Hackage. This might be a good time to do this.

jgm commented 9 years ago

Will do.

+++ Ben Gamari [Feb 16 15 20:08 ]:

It appears that this is already fixed in master but has yet to be pushed to Hackage. This might be a good time to do this.

Reply to this email directly or view it on GitHub: https://github.com/jgm/cheapskate/pull/9#issuecomment-74614530

bgamari commented 9 years ago

@jgm Sure, it would probably be best to wait until the FTP is actually agreed upon (although from what I've heard it seems fairly likely that this will be the case).

The issue here is the use of OverloadedStrings. I agree that this isn't great but in my experience the pain of importing Foldable et al is felt more often than OverloadedStrings is used, so I believe that the trade-off made by FTP is the right one.

What does bother me about the FTP is the fact that the monomorphic forms offered by Data.List vanish, but it seems that little can be done about this.

jgm commented 9 years ago

+++ Ben Gamari [Feb 17 15 06:45 ]:

@jgm Sure, it would probably be best to wait until the FTP is actually agreed upon (although from what I've heard it seems fairly likely that this will be the case).

The issue here is the use of OverloadedStrings. I agree that this isn't great but in my experience the pain of importing Foldable et al is felt more often than OverloadedStrings is used, so I believe that the trade-off made by FTP is the right one.

Ah yes, I'd forgotten I was using that. Well, good. Using an explicit list of Char ['X', 'Y', 'Z'] would be another alternative, then.

Anyway...you indicated that the issue is already resolved by an existing commit. Let me know if that's not the case. I released the latest on Hackage last night.