PythonCall SSL Verify Error when calling pd.read_csv

New to Julia
1

Hello, I’m trying to read a CSV file with Pandas, and I’m getting the following error:

function example52()
    points = pd.read_csv("https://raw.githubusercontent.com/thomasnield/machine-learning-demo-data/master/regression/single_independent_variable_linear_small.csv", delimiter=pystr(",")).itertuples()
end

And the stack trace

ERROR: Python: URLError: <urlopen error [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: unable to get local issuer certificate (_ssl.c:1000)>
Python stacktrace:
 [1] do_open
   @ urllib.request ~/work/MathForDataScience/.CondaPkg/env/lib/python3.12/urllib/request.py:1347
 [2] https_open
   @ urllib.request ~/work/MathForDataScience/.CondaPkg/env/lib/python3.12/urllib/request.py:1392

I know this is probably due to how the Conda environment is configured. I use CondaPkg with no additional configuration, which looks like it uses MicroMamba to do the Python management. I already had to use CondaPkg to downgrade SSL to 3.3.0 so that the Python SSL module could find it, but stuck on this error now.

2

You need to downgrade OpenSSL to 3.0. Julia is currently tracking the OpenSSL LTS.

3

@cjdoris , I am wondering if could consider pinning package versions of some dependencies to match the corresponding dependency versions that Julia uses. That might be able to avoid situations like this.

4

Yep we already do this for libstdc++ here: CondaPkg.jl/src/resolve.jl at 9b73ea5c68bd7e19ba7a946e0ebf96542d0ace4f · JuliaPy/CondaPkg.jl · GitHub

Feel free to make a PR with similar functionality for OpenSSL.