On Dec 29, 2007, at 9:31 , Kristis Makris wrote:
> Would these tests fail if the testsuite is run twice without
> manually resetting the status of the involved bugs in the
bugtracker ?
No, I just have to do manual work, if some tests fail, even there not
always.
I don't understand. If you run the testsuite, and all tests pass,
do you need to do some manual work in the bugtracker in order to
run the testsuite again ?
No, the bug ends up in status 'ASSIGNED', which is the starting
status for the first test.
> I would assume that in order to ensure testing status resolution is
> reliable one would need to dynamically create a bug, set it to the
> expected status, and then run a test case against it (and then
> close the bug). This is certainly quite complicated to do, and
> something we don't have the testing infrastructure to do now.
>
We just need a setup for each bugtracker, don't think it is too
complicated
I would think it is complicated if you wanted to provide a generic
API for manipulating any bugtracker (not just Bugzilla) in order to
setup the testsuite.
Well, it may be a lot of work to be done, but not complicated :-)
You have considered to work with database dumps?
_______________________________________________
scmbug-users mailing list
[email protected]
http://lists.mkgnu.net/cgi-bin/mailman/listinfo/scmbug-users
