A brief skim through these patches looks fine to me.  Just one comment
below.

David

On Mon, Mar 05, 2007 at 09:16:06PM +0100, Eric Kow wrote:
> I'm not sure about changing --disable-ssh-cm to --no-ssh-cm.  It does
> seem to be more consistent with other darcs options, and is probably not
> frequently enough used to sow confusion.
> 
> Mon Mar  5 20:59:02 CET 2007  Eric Kow <[EMAIL PROTECTED]>
>   * Rename --disable-ssh-cm to --no-ssh-cm.
>   
>   This appears to be more consistent with other darcs flags.
>   
> 
> Mon Mar  5 21:00:13 CET 2007  Eric Kow <[EMAIL PROTECTED]>
>   * Add a --ssh-cm flag with --no-ssh-cm as the default.
>   
>   Previously, darcs would launch the ControlMaster by default, but it seems to
>   hang on some large repositories and cause pain.  The user can always add
>   -ssh-cm if s/he wants it on.
>   
> 
> Mon Mar  5 21:07:52 CET 2007  Eric Kow <[EMAIL PROTECTED]>
>   * Flip ssh test to accept a --ssh-cm argument

>                      -- set any global variables
> -                    when (NoSSHControlMaster `elem` os) 
> setSshControlMasterDisabled
> +                    when (NoSSHControlMaster `elem` os || not 
> (SSHControlMaster `elem` os))
> +                      setSshControlMasterDisabled

The DarcsFlag code will (should?) disallow mutually exclusive options, so
we shouldn't need this complicated check here.  We should be able to get by
with:

unless (SSHControlMaster `elem` os) setSshControlMasterDisabled

which I find considerably more readable.

David
_______________________________________________
darcs-devel mailing list
[email protected]
http://lists.osuosl.org/mailman/listinfo/darcs-devel

Reply via email to