Hi Ben and Garrett. I am logged in as the administrator (full access rights). This problem is quite confusing since I recall using a older version of the GHC (standalone) without any incidents. I have no idea how to solve this particular problem. Thanks, ~Caitlin On Fri, Jan 18, 2013 at 11:05 AM, Ben Gamari wrote:

Caitlin writes:

...

Hi all.

I wrote the following test program:

bar x = x * x

on a Win32 (SP3) machine using the Haskell Platform and GHC-7.6.1 (standalone) and I have noticed the same frustrating errors. Any suggestions would be most appreciated.

Are you certain you have write permission for C:\? I'm not terribly with how Windows handles file permissions, but the error seems pretty unambiguous as to its cause. It's a little strange to be placing files like this in the root directory of a volume, anyways. Perhaps try this in a directory which you certainly have adequate permissions in (the Windows equivalent of $HOME perhaps. Wikipedia seems to think this is %UserProfile%).

Cheers,

- Ben

On Fri, Jan 18, 2013 at 11:32 AM, Brandon Allbery wrote:

On Fri, Jan 18, 2013 at 1:05 PM, Ben Gamari wrote:

...

...

on a Win32 (SP3) machine using the Haskell Platform and GHC-7.6.1 (standalone) and I have noticed the same frustrating errors. Any suggestions would be most appreciated.

Are you certain you have write permission for C:\? I'm not terribly with how Windows handles file permissions, but the error seems pretty

I suspect it's more fundamental than that.

...

*** Exception: CreateDirectory ".": permission denied (Access is denied.)

Why is it trying to create "."?

You can specify -o<dir> with a directory that doesn't exist; perhaps GHC always tries to create the output directory? In any case, I can confirm the observed behavior with GHC 7.6.1 on Windows 8, running with elevated privilege. It seems to only be a problem in the root directory tho---while I can observe the problem in both C:\ and D:\, I can't in any subdirectories thereof. /g -- Sent from my mail client.

Ahh...I see. Thanks Brandon. I was unaware of that. ~Caitlin On Fri, Jan 18, 2013 at 1:00 PM, Brandon Allbery wrote:

On Fri, Jan 18, 2013 at 2:53 PM, J. Garrett Morris wrote:

...

...

Why is it trying to create "."?

You can specify -o<dir> with a directory that doesn't exist; perhaps GHC always tries to create the output directory?

In any case, I can confirm the observed behavior with GHC 7.6.1 on Windows 8, running with elevated privilege. It seems to only be a problem in the root directory tho---while I can observe the problem in both C:\ and D:\, I can't in any subdirectories thereof.

That makes a certain amount of sense; the root directory of a drive is treated somewhat specially for backward compatibility reasons. (In particular, "." is emulated --- on FAT, at least, neither "." nor ".." exists in a drive root, and this behavior may well be emulated to keep from confusing older programs.)

-- brandon s allbery kf8nh sine nomine associates allbery.b@gmail.com ballbery@sinenomine.net unix, openafs, kerberos, infrastructure, xmonad http://sinenomine.net

_______________________________________________ Glasgow-haskell-users mailing list Glasgow-haskell-users@haskell.org http://www.haskell.org/mailman/listinfo/glasgow-haskell-users

On Fri, Jan 18, 2013 at 01:05:05PM -0700, Caitlin wrote:

I deleted the Haskell Platform installation, manually removed all traces of GHC and the Hakell Platform from my registry and various folders, then re-installed the Haskell Platform. I created a folder under the 'C:\' drive, copied my .hs files there, changed to that subfolder started ghci. The file loaded without error. This seems like a problem the developers might address in the next release?

Thanks for the report. I've fixed it in HEAD. Thanks Ian