protect against Tk not having `_default_root` attr during event loop

[IanButterworth]

I don't understand why this segfault is happening after the test suite finishes, but this should protect against it?

Test Summary:  | Pass  Total     Time
Foofoofoooo.jl |  614    614  5m39.0s
signal (11): Segmentation fault
in expression starting at none:0
_PyInterpreterState_GET at /usr/local/src/conda/python-3.11.0/Include/internal/pycore_pystate.h:116 [inlined]
get_dict_state at /usr/local/src/conda/python-3.11.0/Objects/dictobject.c:251 [inlined]
new_dict at /usr/local/src/conda/python-3.11.0/Objects/dictobject.c:722 [inlined]
PyDict_New at /usr/local/src/conda/python-3.11.0/Objects/dictobject.c:841
type_ready_set_dict at /usr/local/src/conda/python-3.11.0/Objects/typeobject.c:6143 [inlined]
type_ready at /usr/local/src/conda/python-3.11.0/Objects/typeobject.c:6462 [inlined]
PyType_Ready at /usr/local/src/conda/python-3.11.0/Objects/typeobject.c:6513
_PyObject_GenericGetAttrWithDict at /usr/local/src/conda/python-3.11.0/Objects/object.c:1266
PyObject_GenericGetAttr at /usr/local/src/conda/python-3.11.0/Objects/object.c:1367 [inlined]
module_getattro at /usr/local/src/conda/python-3.11.0/Objects/moduleobject.c:761
PyObject_GetAttr at /usr/local/src/conda/python-3.11.0/Objects/object.c:916
PyObject_GetAttr at /home/ubuntu/.julia/packages/PythonCall/dsECZ/src/cpython/pointers.jl:299 [inlined]
pygetattr at /home/ubuntu/.julia/packages/PythonCall/dsECZ/src/abstract/object.jl:60
getproperty at /home/ubuntu/.julia/packages/PythonCall/dsECZ/src/Py.jl:272 [inlined]
#9 at /home/ubuntu/.julia/packages/PythonPlot/f591M/src/pygui.jl:194
macro expansion at ./asyncevent.jl:281 [inlined]
#666 at ./task.jl:134
unknown function (ip: 0x7ff7e519218f)
_jl_invoke at /cache/build/default-amdci4-2/julialang/julia-release-1-dot-8/src/gf.c:2377 [inlined]
ijl_apply_generic at /cache/build/default-amdci4-2/julialang/julia-release-1-dot-8/src/gf.c:2559
jl_apply at /cache/build/default-amdci4-2/julialang/julia-release-1-dot-8/src/julia.h:1843 [inlined]
start_task at /cache/build/default-amdci4-2/julialang/julia-release-1-dot-8/src/task.c:931
Allocations: 838271008 (Pool: 837751882; Big: 519126); GC: 109

Unfortunately I can't reproduce this segfault locally, but will check this works

[IanButterworth]

Still segfaults but differently

signal (11): Segmentation fault
in expression starting at none:0
_PyInterpreterState_GET at /usr/local/src/conda/python-3.11.0/Include/internal/pycore_pystate.h:116 [inlined]
get_dict_state at /usr/local/src/conda/python-3.11.0/Objects/dictobject.c:251 [inlined]
new_dict at /usr/local/src/conda/python-3.11.0/Objects/dictobject.c:722 [inlined]
PyDict_New at /usr/local/src/conda/python-3.11.0/Objects/dictobject.c:841
type_ready_set_dict at /usr/local/src/conda/python-3.11.0/Objects/typeobject.c:6143 [inlined]
type_ready at /usr/local/src/conda/python-3.11.0/Objects/typeobject.c:6462 [inlined]
PyType_Ready at /usr/local/src/conda/python-3.11.0/Objects/typeobject.c:6513
_PyObject_GenericGetAttrWithDict at /usr/local/src/conda/python-3.11.0/Objects/object.c:1266
PyObject_GenericGetAttr at /usr/local/src/conda/python-3.11.0/Objects/object.c:1367 [inlined]
module_getattro at /usr/local/src/conda/python-3.11.0/Objects/moduleobject.c:761
PyObject_GetAttr at /usr/local/src/conda/python-3.11.0/Objects/object.c:916
PyObject_GetAttr at /home/ubuntu/.julia/packages/PythonCall/dsECZ/src/cpython/pointers.jl:299 [inlined]
macro expansion at /home/ubuntu/.julia/packages/PythonCall/dsECZ/src/Py.jl:131 [inlined]
pyhasattr at /home/ubuntu/.julia/packages/PythonCall/dsECZ/src/abstract/object.jl:37
#9 at /home/ubuntu/.julia/packages/PythonPlot/ei8cR/src/pygui.jl:194
macro expansion at ./asyncevent.jl:281 [inlined]
#666 at ./task.jl:134
unknown function (ip: 0x7fac1b57080f)
_jl_invoke at /cache/build/default-amdci4-2/julialang/julia-release-1-dot-8/src/gf.c:2377 [inlined]
ijl_apply_generic at /cache/build/default-amdci4-2/julialang/julia-release-1-dot-8/src/gf.c:2559
jl_apply at /cache/build/default-amdci4-2/julialang/julia-release-1-dot-8/src/julia.h:1843 [inlined]
start_task at /cache/build/default-amdci4-2/julialang/julia-release-1-dot-8/src/task.c:931
Allocations: 860081766 (Pool: 859557939; Big: 523827); GC: 110

[stevengj]

Why is the gui even being initialized during the test suite? We set the backend to Agg…

[IanButterworth]

We run our test suite with xvfb because it also includes GLMakie plots. Could it be that?

[IanButterworth]

I should add that this didn't happen with PyPlot.jl

[stevengj]

https://github.com/stevengj/PythonPlot.jl/blob/9a7d10e72b8800864ebabe4e8cbbae6b93acfa4b/test/runtests.jl#L1 should take precedence unless you have already loaded/initialized PythonPlot.jl somewhere else