Using pandas in the importers

[Rishabh Vaid]

Hi folks,

I'm trying to get set up with beancount in WSL. The desire to use beancount is actually what motivated me to install WSL and start using the command line. As a result, I am running into some nooby issues, would be grateful if anyone here could help me out!

I'm using pandas inside my importers to read in the csv file and parse it. However, I keep getting "No module named 'pandas'" errors when I try to run bean-identify. I have pip installed pandas in my virtual env, but not clear if I need to install it somewhere else.

Thanks for your help!

[Rishabh Vaid]

Update: I pip installed pandas outside the venv and now it worked. I'm not completely sure why though, and whether this is the correct way to do things. Does anyone have any pointers?

[nug get]

Hi

this smells like confusion with the python interpreters. You may want to debug through your bean-commands and check whether the correct python executable is uses all the time. Also make sure the correct venv is activated all the time.
Good luck! For me, WSL works very well.

best

[redst...@gmail.com]

The interpreter/venv issue aside, you may want to check out petl which is likely more specialized than pandas for operations around importing csvs into beancount.