$ darcs get --partial http://darcs.haskell.org/ghc
This is the GHC darcs repository (HEAD branch)

For more information, visit the GHC developer wiki at
  http://hackage.haskell.org/trac/ghc
**********************
Copying patch 210 of 210... done.
Applying patch 42 of 209... darcs.exe: bug in darcs!
Impossible case at PatchRead.lhs:58 compiled 23:01:08 Jun  4 2007
Please report this to bugs@darcs.net,
If possible include the output of 'darcs --exact-version'.

$ darcs --exact-version
darcs compiled on Jun  4 2007, at 23:02:57
# configured Mon Jun  4 22:59:17 USMST 2007
./configure --disable-mmap

Context:

[TAG 1.0.9
Tommy Pettersson <ptp@lysator.liu.se>**20070603213706] 

This material has been prepared by individual sales and/or trading personnel and does not constitute investment research.  Please follow the attached hyperlink to an important disclaimer: http://www.credit-suisse.com/emea/legal

==============================================================================
Please access the attached hyperlink for an important electronic communications disclaimer: 

http://www.credit-suisse.com/legal/en/disclaimer_email_ib.html
==============================================================================

This issue has been resolved, it was the network generating a bad patch.
Perhaps a sanity check on the patches would be a good idea?

	-- Lennart

Thanks for that.  There is a new hashed inventory format in the current stable
branch, and it seems like that would help us implement the right sanity checks.

David, would it make sense to add --verify-hashes as an option to darcs pull?

Actually, I think it makes more sense to remove the --verify-hashes option
altogether.  For the moment, darcs always checks hashes in any case, regardless
of that flag, and I think I'd prefer to continue this behavior.  We're too
easily bitten by network corruption for me to be really comfortable without this
safety net.  And it doesn't seem that verifying the hashes is really a bottleneck.