2008/5/22 Eric Kow <[EMAIL PROTECTED]>:
> Do you know what sequence of darcs command(s) triggered this?  From what 
> you're
> saying, we should be able to solve this with a simple rollback, right?
> Hopefully we can generate a test case for Pekka.
>
> I'll +droundy when we've settled this basic triage-y stuff (unless he's 
> already
> paying attention)

Oops. While I could not figure out how it could happen, culprit is
kind a obvious. dontShiftLast stuff in diffArr does not check that
bits both in c_a and c_b are True when changing them to False.

The following patch moves the end-of-file check to shiftBoundaries.

I can not compile unit tests for shiftBoundaries,  so they might fail.
Does they need ghc 6.6? Some extra package?

-- 
Pekka.Pessi mail at nokia.com

Attachment: making-sure-hunk-is-not-shifted-beyond-end-of-file-in-shiftboundaries.dpatch
Description: Binary data

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

Reply via email to