Tue Apr 7 18:58:39 EDT 2009 cognomi...@gmail.com
* show modifier
Ignore-this: a8d514eb536eb46ca4680f10c8316333
M ./Yi/Event.hs -1 +2
M ./Yi/Keymap/Emacs.hs +4
--~--~-~--~~~---~--~~
Yi development mailing list
yi-devel@googlegroups.com
http://groups.g
I agree that this is too slow. I'm on a very fast machine, and when GHC is
hot yi starts up pretty quickly (though I still have time to see the
"Recompiling custom yi" message). However, when GHC is cold it takes several
seconds to start up.
I would much prefer simply adding documentation (perhaps
GHC actually allows import cycles if you use hs-boot files and {-# SOURCE
#-} pragmas.
http://www.haskell.org/ghc/docs/latest/html/users_guide/separate-compilation.html#mutual-recursion
-Kevin Ballard
On Fri, Apr 10, 2009 at 2:54 PM, Stéphane Payrard wrote:
> Resend of the select-all in a buffe
On Fri, 2009-04-10 at 18:20 +0200, Jean-Philippe Bernardy wrote:
> I don't think it's wise to make "how to load the configuration"
> configurable. We already have a compile-time flag to bypass
> configuration anyway (hacking).
Yeah, I think you're right . . . that would be a bit silly.
> At the
> Hmm, my primary machine is very old, so this causes a noticeable startup
> delay. I tried this on my Mac and noticed it was much more acceptably
> quick. Perhaps this could be a preference?
I don't think it's wise to make "how to load the configuration"
configurable. We already have a compile-t
On Fri, 2009-04-10 at 09:05 +0200, Jean-Philippe Bernardy wrote:
> Note that GHC should not actually recompile: just figure out
> dependencies are not changed then quit.
Yep.
> The advantage is that we rely on GHC to figure out dependencies.
> Hence, we can allow multi-module configurations, et
On Thu, Apr 9, 2009 at 11:55 PM, Jeff Wheeler wrote:
>
> On Wed, Apr 8, 2009 at 5:03 AM, Jean-Philippe Bernardy
> wrote:
>
>> This is experimental: please shout if you think this is too slow.
>
> Hmm, I think I do. Previously, Yi launched pretty much instantly, and
> now this takes about a seco
