Thanks for the application + analysis :-). In 20/20 hindsight I shouldn't have removed the svn:hold property while while debugging the test.
Neels J Hofmeyr wrote on Sat, Aug 20, 2011 at 02:51:32 +0200: > Daniel, > > I have since grasped that all you were doing was helping out with coding a > regression test, applied your patch to the "working" 'hold' branch WC and > ran the test. XFails :) And then realized the error: > > On 08/18/2011 06:29 PM, Daniel Shahaf wrote: > > I used a changelist for debugging purposes during test development: > > > > - svn propset svn:hold A/D/G/pi > > - svn commit A/D/G > > - XFAIL > > + svn cl foo A/D/G/rho > > + svn commit A/D/G --cl foo > > + XPASS > > > > When you applied the changelists code, you removed the line that set the > svn:hold prop. So your passing test expects there to be *no* svn:hold > property on the file. In the real test, there *is* an svn:hold prop and the > test fails because it expects none. If I add an svn:hold prop to > expected_disk, the test XPASSes on the 'hold' branch. :D > > ~Neels > >
