Looks more like a cygwin bug to me. rm -rf isn't recursing into the subdirectories, and somehow is running afoul of a path-length restriction. It's not actually a test that's failing, but the cleanup after a test. I suppose we could work around this cygwin bug by doing the recursive rm manually, but that seems overly stupid.
David On Tue, Sep 23, 2008 at 7:06 AM, Eric Kow <[EMAIL PROTECTED]> wrote: > Hi David, > > Looking at > > http://buildbot.darcs.net/builders/zooko%20allmydata%20virtual2%20Windows-XP%20i386/builds/79/steps/test/logs/stdio > > it seems like with the recent conflict-doppleganger translation, we've > run afoul of Windows filename length restrictions... not darcs > specifically, but the test trying to remove a directory (unless I'm > misreading things). Any thoughts? Implications for darcs-on-Windows > outside of this test? > > Thanks! > > -- > Eric Kow <http://www.nltg.brighton.ac.uk/home/Eric.Kow> > PGP Key ID: 08AC04F9 > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.6 (GNU/Linux) > > iD8DBQFI2M3IBUrOwgisBPkRAkFXAKDMpJ/+GbwGuXYbU0ZyEkFuXdaDRACeP9kK > Im5wDnsFEhTpMPnnBlu/pEU= > =vktA > -----END PGP SIGNATURE----- > > _______________________________________________ darcs-users mailing list [email protected] http://lists.osuosl.org/mailman/listinfo/darcs-users
