I found that I can reproduce this from a clean system. I'm not sure why I didn't get this error when I ran the tests locally, but I can now. Maybe my system was too dirty? Anyway, I bisected it, and this is a regression caused by:
elopio@calchas:~/workspace/canonical/snapcraft$ git bisect bad
fdab1bbab5ed6f17cc7007195bca9946b54e103d is the first bad commit
commit fdab1bbab5ed6f17cc7007195bca9946b54e103d
Author: Sergio Schvezov <email address hidden>
Date: Thu Feb 11 00:54:03 2016 -0300
Replacing the need for pyversions
pyversions creates a chicken and egg problem, so we use globbing
to work around that.
Sergio: why did you mark it as released again?
I found that I can reproduce this from a clean system. I'm not sure why I didn't get this error when I ran the tests locally, but I can now. Maybe my system was too dirty? Anyway, I bisected it, and this is a regression caused by:
elopio@ calchas: ~/workspace/ canonical/ snapcraft$ git bisect bad 7cc7007195bca99 46b54e103d is the first bad commit 7cc7007195bca99 46b54e103d
fdab1bbab5ed6f1
commit fdab1bbab5ed6f1
Author: Sergio Schvezov <email address hidden>
Date: Thu Feb 11 00:54:03 2016 -0300
Replacing the need for pyversions
pyversions creates a chicken and egg problem, so we use globbing
to work around that.
LP: #1541451
Signed-off-by: Sergio Schvezov <email address hidden>