Presumably --partial should be implicit when the source repository is partial.

Sample output:
Copying patch 134 of 1356...darcs-1.0.6-static-linux-i386:
/media/UDISK_2_0/taps/_darcs/patches/20050816153639-31b3d-251d3fabe9f0521776f55277e8c51c86b4cb00ea.gz:
openBinaryFile: does not exist (No such file or directory)

 ~/darcs-1.0.6-static-linux-i386 get --partial /media/UDISK_2_0/taps
Copying patch 133 of 133... done!
Applying patch 132 of 132... done.
Finished getting.

It would probably be wise to report this to the user, something like "the
repository we're getting is partial, so assuming --partial"

On Tue, Sep 05, 2006 at 01:43:00PM +0000, Eric Kow wrote:
> It would probably be wise to report this to the user, something like "the
> repository we're getting is partial, so assuming --partial"

I agree, this would be a good idea.  I'm not sure how easy this will
be, as we don't know that a repo is partial until we've downloaded the
files.  It won't be *too* hard, but probably isn't appropriate for a
haskell newbie.

David

Should this ticket and its duplicates be considered "resolved-in-unstable" since
switching to "Darcs 2" and the darcs-2 repo format avoids the issue?

On Wed, Jan 09, 2008 at 04:43:48AM -0000, Mark Stosberg wrote:
> Should this ticket and its duplicates be considered "resolved-in-unstable" since
> switching to "Darcs 2" and the darcs-2 repo format avoids the issue?

Hmmm.  I suppose so.  Although in practice one need only switch to darcs 2
and the hashed format to avoid the issue.
-- 
David Roundy
Department of Physics
Oregon State University

Just noting that if we were to re-open this bug, then we should probably
implement some mechanism for detecting partial repos (see issue1318) first.

But I'm assuming we can leave this closed for now (sorry, Karel) because partial
repositories are moribund.  Do please feel free to argue the case.