Is it okay to install pyodbc module with PIP to a cloned environment?

[alex_friant]

I noticed when I was trying to install the pydobc ver 4.0.30 module using the Python Package Manager interface (in ArcGIS Pro 2.5) that pyodbc would never really install. There were no error messages or anything, it would just not install.

To see what was going on, I tried doing it using conda from the command line, but it would warn me that the gdal module would get rolled back:

I said no.

My next move was to try it using pip, and it installed without any warnings or anything:

My questions are:

• Is this a good practice?
• Am I potentially going to cause issues doing this?

I'll try it out and see what happens, I was just wondering if any python gurus out there could shed light on potential issues I could expect from using PIP to bypass warnings from Conda.

Thanks!

[DanPatterson_Retired]

Avoid pip if it is available within the conda environment

conda install blah_blah --no-pin --dry-run

# check things out and if you thing nothing serious will happen

conda install blah_blah --no-pin


# if things go really, really bad

conda list --revisions

# then pick the revision when things were better

conda install --revision [revision number]‍‍‍‍‍‍‍‍‍‍‍‍‍‍