Closed artzet-s closed 2 years ago
Sure. This is a known behavior (not a bug).At the end of the process of the viewer, there is a seg. fault on Linux. If you run the viewer on IPython with qt (%gui qt), it will works I think.
Thanks @pradal you are right, is work's fine with ipython (%gui qt). Otherwise, I saw by testing, that PIL dependency is missing in CI.
If it works only with ipython, should we add ipython in the dependency ? Can we use nosetest with ipython to test this ? (Otherwise this test will fail each time in a desktop version)
Nevertheless, I not convinced by "not a bug", for me segfault = bug by definition.
I think we need to not launch the viewer in the unit tests. No, No ipython deps. The seg fault is at the end of the process. So ,even if it is not nice, it does not impact the use of PlantGL.
Perhaps @fredboudon knows how to fix it. It is related to the viewer.
Le 16/11/2017 à 15:15, artzet-s a écrit :
Thanks @pradal https://github.com/pradal you are right, is work's fine with ipython (%gui qt). Otherwise, I saw by testing, that PIL dependency is missing in CI.
If it works only with ipython, should we add ipython in the dependency ? Can we use nosetest with ipython to test this ? (Otherwise this test will fail each time in a desktop version)
Nevertheless, I not convinced by "not a bug", for me segfault = bug by definition.
— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/openalea/plantgl/issues/15#issuecomment-344934832, or mute the thread https://github.com/notifications/unsubscribe-auth/ABEeShRT2KHxaTrHPcVGclrN8XAY4S_Wks5s3EN1gaJpZM4Qgb7X.
Tested on a new ubuntu 14 desktop VM with a plantgl installation following .travis.ci Tested with
nosetests test_viewer.py
andpython test_viewer.py
, same result.@fredboudon @pradal