[darcs-devel] darcs patch: Rename AtExit module to Global. (and 2
more)
Eric Y. Kow
eric.kow at gmail.com
Thu Aug 17 03:31:05 PDT 2006
Hi,
On Thu, Aug 17, 2006 at 11:59:50 +0200, Tommy Pettersson wrote:
> They conflict with an earlier patch of yours not yet applied to
> unstable, changing FilePath to Maybe FilePath in some input /
> output redirections. (It was not hard to resolve.) Do we want
> that patch too, i.e., does it fix anything?
No, it's ok. I deliberately submitted the patch this way so that it
wouldn't depend on the other patch being accepted. (I should have
pointed that out... oops).
I have a new patch that depends on this one, but does the same FilePath
to Maybe FilePath thing.
> Btw, I get a bad feeling about this Global thing. I'm worried
> we're introducing ugly hacks just so we can solve problems _our_
> way and not the Haskell way.
Yeah :-) I'm not sure which is the right way either. It seems like our
choices (so far) are
1) clean/correct but messy (passing parameters)
By messy, I mean, we'd have to add an extra ssh-cm-disabled (or
DarcsFlag) parameter to these functions and maybe more
runSSH, execSSH
copyFileOrUrl, copySSH
gzFetchFilePS, fetchFilePS
read_repo, lazily_read_repo, read_repo_private
or
2) ugly/hacky + tidy (unsafe IORefs)
By tidy, I'm only talking about not adding those extra parameters
> But then I'm just beginning to really understand monads, so I'm not
> qualified to say what is and what's not good Haskell.
Hmm... I wonder if anybody feeling more confident than both of us
would care to comment, because I personally could go both ways
(leaning towards #2 as the tidier one, but not sure if it's
ultimately wrong-headed)
Cheers,
--
Eric Kow http://www.loria.fr/~kow
PGP Key ID: 08AC04F9 Merci de corriger mon français.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 186 bytes
Desc: not available
Url : http://lists.osuosl.org/pipermail/darcs-devel/attachments/20060817/b5ea58de/attachment.pgp
More information about the darcs-devel
mailing list