Skip to content

Latest commit

 

History

History
43 lines (32 loc) · 3.98 KB

best_python_jupyter_pycharm_experience.md

File metadata and controls

43 lines (32 loc) · 3.98 KB

Best Python/Jupyter/PyCharm/VSCode/Atom experience + report generation with Pandoc filters

(backup of the reddit post)

I'm going to share with you my favourite way of using Python and Jupyter for research and reports. And it's not simply using Jupyter. I want to write code in PyCharm and use Pandoc filters for reports.

Example

There is an example (from here) that can be edited in PyCharm, interactively run in Atom+Hydrogen and exported via Pandoctools.

VS Code

Or you can use VS Code instead of PyCharm. Everything seems to work out of the box except you need to install pylint to each Python env you are going to use. If you would run Jupyter in Atom/Hydrogen then you can disable Jupyter functionality in VS Code: "Data Sciense: Enabled" (this would hide annoying "run cell" inserts).

Atom

Or you can do everything in Atom Editor. For this you need to install ide-python (+ enable pylint there), install python-tools, install python-language-server to each working Python environment and tune pylint globally: edit or create ~/.pylintrc file:

[MESSAGES CONTROL]
disable=bad-continuation,missing-docstring,pointless-string-statement,invalid-name,too-many-locals,too-many-arguments
[FORMAT]
max-line-length=119

Create it on Windows:

cd /d %UserProfile%
echo bla > .pylintrc