IPython crash when error occurs after import tyjuliacall julia module

[songjhaha]

Python's original repl works fine

In [1]: from tyjuliacall import Main

In [2]: Main.eval(Main.Meta.parse("struct Point x end"))

In [3]: a = Main.Point(1)

In [4]: a*a # try to
ERROR:root:Internal Python error in the inspect module.
Below is the traceback from this internal error.

Traceback (most recent call last):
  File "/home/songjhaha/miniconda3/envs/dev/lib/python3.7/site-packages/IPython/core/interactiveshell.py", line 3553, in run_code
    exec(code_obj, self.user_global_ns, self.user_ns)
  File "<ipython-input-4-ac686954dc0d>", line 1, in <module>
    a*a
  File "/home/songjhaha/miniconda3/envs/dev/lib/python3.7/site-packages/tyjuliasetup/jv.py", line 64, in __mul__
    return __jl_mul__(self, other)
RuntimeError: MethodError: no method matching length(::Point)
Closest candidates are:
  length(!Matched::Union{Base.KeySet, Base.ValueIterator}) at ~/.jnumpy/julias/julia-1.7/share/julia/base/abstractdict.jl:58
  length(!Matched::Union{LinearAlgebra.Adjoint{T, S}, LinearAlgebra.Transpose{T, S}} where {T, S}) at ~/.jnumpy/julias/julia-1.7/share/julia/stdlib/v1.7/LinearAlgebra/src/adjtrans.jl:171
  length(!Matched::Base.InvasiveLinkedListSynchronized) at ~/.jnumpy/julias/julia-1.7/share/julia/base/task.jl:604
  ...
Stacktrace:
 [1] _similar_shape(itr::Point, #unused#::Base.HasLength)
   @ Base ./array.jl:600
 [2] _collect(cont::UnitRange{Int64}, itr::Point, #unused#::Base.HasEltype, isz::Base.HasLength)
   @ Base ./array.jl:655
 [3] collect(itr::Point)
   @ Base ./array.jl:649
 [4] broadcastable(x::Point)
   @ Base.Broadcast ./broadcast.jl:704
 [5] broadcasted(::Function, ::Point, ::Point)
   @ Base.Broadcast ./broadcast.jl:1300
 [6] jl_mul(self::TyPython.CPython.Py, other::TyPython.CPython.Py)
   @ TyJuliaSetup ~/miniconda3/envs/dev/lib/python3.7/site-packages/tyjuliasetup/src/TyJuliaSetup.jl:320
 [7] jl_mul(self::Ptr{TyPython.CPython.PyObject}, _vectorargs::Ptr{TyPython.C.Ptr{TyPython.CPython.PyObject}}, argc::Int64)
   @ TyJuliaSetup ~/miniconda3/envs/dev/lib/python3.7/site-packages/jnumpy/TyPython/src/CPython.Dev.jl:90
 [8] jl_mul(self::Ptr{TyPython.CPython.PyObject}, _vectorargs::Ptr{TyPython.C.Ptr{TyPython.CPython.PyObject}}, argc::Int64)
   @ TyJuliaSetup ~/miniconda3/envs/dev/lib/python3.7/site-packages/jnumpy/TyPython/src/CPython.Dev.jl:111

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/songjhaha/miniconda3/envs/dev/lib/python3.7/site-packages/IPython/core/interactiveshell.py", line 2099, in showtraceback
    stb = value._render_traceback_()
AttributeError: 'RuntimeError' object has no attribute '_render_traceback_'
...

[songjhaha]

Maybe it's the problem of my IPython. After import tyjuliacall.Main, then try to trigger some error, IPython crashes.

same error message:

In [1]: from tyjuliacall import Main

In [2]: a
ERROR:root:Internal Python error in the inspect module.
Below is the traceback from this internal error.

Traceback (most recent call last):
  File "/home/songjhaha/miniconda3/envs/pycall/lib/python3.7/site-packages/IPython/core/interactiveshell.py", line 3553, in run_code
    exec(code_obj, self.user_global_ns, self.user_ns)
  File "<ipython-input-2-3f786850e387>", line 1, in <module>
    a
NameError: name 'a' is not defined

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/home/songjhaha/miniconda3/envs/pycall/lib/python3.7/site-packages/IPython/core/interactiveshell.py", line 2099, in showtraceback
    stb = value._render_traceback_()
AttributeError: 'NameError' object has no attribute '_render_traceback_'
...

[thautwarm]

The support for IPython is very limited. Shall we just remove the following line?

tyjuliacall/tyjuliasetup/jv.py

Line 138 in 94d8245

def _repr_pretty_(self, p, cycle):

[thautwarm]

A simple gooling shows that this might not be related to tyjuliacall. It's related to virtualenv.

microsoft/vscode-jupyter#9306

microsoft/vscode-jupyter#9306 (comment)

Your python environment is kinda mismatched. You have Python 3.8 with a much newer IPython. They're likely not compatible.

[songjhaha]

that's true, a new env with python 3.8 works fine