On Sun, Jan 28, 2007 at 07:43:12PM +0100, Eric Y. Kow wrote: > > Thu Jan 25 07:38:03 PST 2007 Tommy Pettersson <[EMAIL PROTECTED]> > > * add test for replace that messes with unrecorded hunks > > > echo "x" > foo > > darcs rec -am x > > echo "y" > foo > > darcs replace x y foo > > > > # this fails > > echo "hej" > foo > > darcs rec -am hej > > echo "hopp" > foo > > darcs replace hej hopp foo > > Thanks for the test! > > If I'm reading this right, these are the same test but just repeated so > that we can slip into the condition where it fails, right?
I didn't think about --ignore-times when I wrote the test. I expected the first replace to fail, and when it didn't and I finally succeeded in getting a failure by adding a second replace I just left everything in because I figured it would increase the probability of catching bugs. :-) Now when I understand more what was happening I think the test (at least the comments) should be updated. Btw, if the opts Repository refactoring patch is going to be amended, there are some fixes for list_authors and make_changelog I sent in another patch that are better incorporated in the refactoring patch itself, possibly in another way if passing [] to identifyRepository is abandoned. -- Tommy Pettersson <[EMAIL PROTECTED]> _______________________________________________ darcs-devel mailing list [email protected] http://lists.osuosl.org/mailman/listinfo/darcs-devel
