Re: [SyncEvolution] Concurrent sync sessions update - Almost done

2012/3/15 Chris Kühl <blixtra(a)gmail.com&gt;: > On Wed, Mar 14, 2012 at 11:28 AM, Patrick Ohly <patrick.ohly(a)intel.com&gt; wrote: >> On Thu, 2012-03-08 at 17:18 +0100, Chris Kühl wrote: >>> It's been rebased and we've squashed the changes into about a dozen >>> commits. The tests seem to give us the same results as before the >>> rebase. >> >> And those are? In other words, which tests are known to fail, and why? >> > > I've pushed a branch called css-without-known-test-failures[1] whose > most recent commit comments out all the tests in test/test-dbus.py > that fail. However, there as a bug I introduced during the last > rebase. The fix for that can be found here[2]. > > All but one of the failures is due to the incomplete AutoSync and > Connection functionality. The restart test has never worked for us on > any branch. That test should work without problems if you run test-dbus.py while being in the same directory as syncevo-dbus-server binary. That is because execve's first parameter should be a path (relative or absolute) to binary and in case of test-dbus.py only 'syncevo-dbus-server' is passed here. It is not absolute path and thus execlv looks for binary in current running directory only. Seems like execve does not look for binary in PATH environment variable.