Python/Matplotlib

[nilskvilvang]

I just started looking at Quarto, thinking I might switch from RMarkdown (at some point). When running through the example code at the quarto web page, I can run the R and Julia examples. Still, when trying to run the python code chunk, I get this: "ModuleNotFoundError: No module named 'matplotlib'" - I have the latest Python (3.10). I run the latest RStudio (2022.07.1 Build 554).

[chrisvwx]

I guess you could try python3 -m pip install matplotlib though I am not a python guru at all, and this could be wrong.

I am getting a similar error to you (ModuleNotFoundError: No module named 'nbclient') even though a pip command (python3 -m pip install nbclient) says that the module in question is already installed. I wonder if I somehow have a python version mismatch.

[chrisvwx]

Thanks for the suggestion. My output (below) looks similar to yours, though my focus is on Julia, and I don't use R. To be clear, the basic Python-based Quarto "hello" test doc on the Quarto home page works for me, its just the Julia-based doc that fails on me.

Here's the output of quarto check:

[✓] Checking Quarto installation......OK
      Version: 1.1.251
      Path: /Applications/quarto/bin

[✓] Checking basic markdown render....OK

[✓] Checking Python 3 installation....OK
      Version: 3.10.8
      Path: /usr/local/opt/[email protected]/bin/python3.10
      Jupyter: 5.0.0
      Kernels: julia--1.1, julia-1.7, julia-0.6, julia-1.5, julia-1.2, python3

[✓] Checking Jupyter engine render....OK

[✓] Checking R installation...........(None)

      Unable to locate an installed version of R.
      Install R from https://cloud.r-project.org/

[chrisvwx]

Just updated my Quarto version so that quarto check says that my Quarto version is 1.2.269, with the same failure when trying to compile the basic Julia file. I.e. with the test Julia file in jtest.qmd, the command quarto render jtest.qmd --to html gives the following output:

Starting julia-1.7 kernel...Traceback (most recent call last):
  File "/Applications/quarto/share/jupyter/jupyter.py", line 21, in <module>
    from notebook import notebook_execute, RestartKernel
  File "/Applications/quarto/share/jupyter/notebook.py", line 17, in <module>
    from nbclient import NotebookClient
ModuleNotFoundError: No module named 'nbclient'

[rgaiacs]

@christianpeel Thanks for the extra information. From your output, you are using the Jupyter kernel named julia-1.7. You must install the nbclient in the julia-1.7 kernel. Try

$ conda activate julia-1.7
$ conda install nbclient

[chrisvwx]

It looks like the only conda I have is in ~/.julia/conda that was possibly installed as a part of IJulia. I see that there is a Julia package Conda.jl which was not immediately helpful. The Julia conda wanted to be initialized (conda init bash) after which conda activate julia-1.7 says it can't find julia-1.7:

EnvironmentNameNotFound: Could not find conda environment: julia-1.7
You can list all discoverable environments with `conda info --envs`.

I'm not sure I believe that, since the command quarto render jtest.qmd now gives output that starts by saying it is 'starting the julia-1.7 kernel':

Starting julia-1.7 kernel...Traceback (most recent call last):
  File "/Applications/quarto/share/jupyter/jupyter.py", line 21, in <module>
    from notebook import notebook_execute, RestartKernel
  File "/Applications/quarto/share/jupyter/notebook.py", line 17, in <module>
    from nbclient import NotebookClient
  File "/Users/cpeel/.julia/conda/3/lib/python3.7/site-packages/nbclient/__init__.py", line 6, in <module>
    from .client import NotebookClient, execute  # noqa: F401
  File "/Users/cpeel/.julia/conda/3/lib/python3.7/site-packages/nbclient/client.py", line 18, in <module>
    from traitlets import (
ImportError: cannot import name 'Callable' from 'traitlets' (/Users/cpeel/.julia/conda/3/lib/python3.7/site-packages/traitlets/__init__.py)

I don't know that this is a quarto problem, and it's definitely not a Matplotlib problem as described in this thread. Maybe I should go to the Julia slack rather than having you solve Julia problems

[kdheepak]

I was able to work around this issue by installing jupyter, nbclient etc in the conda installation from Julia.

using Pkg;
Pkg.add("Conda")

using Conda
Conda.add("ipython") # probably not required
Conda.add("jupyter")
Conda.add("nbclient")