[ANNOUNCE] GHC 8.0.1 release candidate 2

[Ben Gamari]

Sven Panne <svenpanne at gmail.com> writes:

> 2016-02-16 10:56 GMT+01:00 Herbert Valerio Riedel <hvriedel at gmail.com>:
>
>> [...] but `sig(nature)s` has a precedent, so using `-sigs` wouldn't
>> introduce anything new.
>>
>
> I'm fine with "sigs", my point was only the fact that non-abbreviated words
> seem to be much more common in the flags names (and are easier to
> remember). IMHO it doesn't really matter if the flag names are long: One
> probably doesn't type them on the command line often, they typically live
> in .cabal files, .travis.yml and pragmas where you type them once.
>
> Well... the  -Wnoncanonical-*-instances flag family was the best I could
>> come up with which is reasonably self-descriptive... do you have any
>> better suggestions?
>>
>
> No, and I actually like the long names, see above. :-)

I don't have an opinion here. Especially with tab completion these names
are rarely typed anyways.

>> Fwiw, `ghc --show-options | grep binding` come ups empty
>>
>
> Then the docs are out-of-sync:
> http://downloads.haskell.org/~ghc/master/users-guide/using-warnings.html#ghc-flag--Wlazy-unlifted-bindings
Indeed it appears that this warning was supposed to be removed in 7.10.

I've gone ahead and finished this in https://phabricator.haskell.org/D1922

Cheers,

- Ben
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 472 bytes
Desc: not available
URL: <http://mail.haskell.org/pipermail/ghc-devs/attachments/20160216/dc9268de/attachment.sig>