Am 17.12.2016 um 10:46 schrieb Marion & Christophe JAILLET:
Proposed fix in r1774728.
A solution, stating that the tests have been skipped because of sed
location, would be better, though.
I switched the test to using a simple perl script instead of sed, so
that we have no platform dependencies any more. It was only using the
search-and-replace feature of sed, which is trivial in perl too.
I did not change the failing tests to TODO though, because I'm not sure
about the original intention.
Le 16/12/2016 à 21:46, Jacob Champion a écrit :
On 12/16/2016 12:44 PM, William A Rowe Jr wrote:
I expect this is a bug in the 2.4.x branch. However, I found the same
exceptions in 2.4.23 tag, so this is not a regression, and Joe is still
researching what is happening here, so it doesn't seem to be something
we would want to hold up a release for.
Yeah, in my case I think it's just that the tests don't exit after
skipping (i.e. 48 out of 24 tests are run...). Not a showstopper.
Regards,
Rainer
