Lisa Anders - questionable.quarto

Positron is a new IDE developed by Posit. I’ve been putting off giving it a deep dive. I love RStudio. Maybe it’s the green icons, or the simple 4 box layout, or even just that it’s what I’ve been using for so long but it really feels like a happy place to me.

I was the same way with RMarkdown. You could have torn it from my cold dead hands, but eventually that feature comes along that makes going through the process of changing worth it. And for me with Quarto that was lightbox with its ability to magically resize images. Such a simple thing, such a challenging thing to do outside of Quarto.

First impressions

Changing the theme made it feel like less of a sterile and “other” place. I’m a big fan of dark modes and I actually really liked the built in dark mode.

gear -> theme -> positron dark

There are some good resources for porting your own theme also: https://www.r-bloggers.com/2025/05/porting-my-favorite-rstudio-color-theme-to-positron/

Viewing data

My extensions

Since Positron is a fork of VS Code it relies on extensions for adding various features. Here are some of my favorites:

Air (R Language Support)
Posit Publisher
Black Formatter
Gitlive
Pyright
Jupyter
Jupter Cell Tags
Jupyter Keymap
Jupyter Slide Show
Project Manager
Python Debugger
Quarto
Ruff
Shiny
Scope to this
vscode-pets

Remote connections with SSH

This is bundled with Positron and there’s no need to install anything.

Resource: https://positron.posit.co/remote-ssh.html

LLM integrations

Stay tuned for Posit::Conf 2025!

Running a python project

Reference: https://positron.posit.co/start.html

I’m a big fan of using the git integration to clone down a project, for example this set of demo examples.

We’ll want to make sure that ipykernel is installed:

python3 -m pip install ipykernel

We can discover which python versions we have access to with:

ls /opt/python/

Select the python interpreter

Manage the interpreter in the session (virtual environment) through the top right icon:

Reference: https://positron.posit.co/managing-interpreters.html

You can also manually select the python interpreter with ctrl-shift-p and then Python: Select Interpreter command and select the Enter interpreter path... option.

This gives you the ability to refresh the visible interpreters with the circle arrow symbol.

Figure 3: Python interpreter manual selection

We can create a new interpreter (and then select it through the dropdown) either with uv or pip.

The uv way

Use uv. It will detect that this is a project and create the venv for us when we go to run the application.

Run the application:

uv run app.py
uv run shiny run --reload app.py

The pip way

Setup the venv environment:

python -m venv env
. env/bin/activate
# source env/bin/activate
pip install --upgrade pip
pip install -r requirements.txt

I recommend using env instead of .venv because quarto will automatically pick it up.

Run the application:

shiny run --reload app.py

Register new interpreter as kernel (likely optional)

# Register as kernel
pip install ipykernel jupyter
# python -m ipykernel install --name "my-new-project_venv" --user 
python -m ipykernel install --name "shiny-income-share" --user

Running and previewing an app

Open the folder so that the root directory is the content example you want to run.

Make sure the appropriate virtual environment has been created (following the steps above) and selected as the interpreter.

Click the play button to launch the application.

Creating a new project

The pip way

Create a venv to isolate environments and generate a requirements.txt file so that only the minimum packages needed to support your project are included. Read this for more on Python package management.

python3 -m pip freeze > requirements.txt
python3 -m pip install -r requirements.txt

The uv way

Create a manifest.json file to support git-backed publishing

For uv projects, they can be initialized and updated with (change format and packages as needed):

    uv init --app --python 3.12.6
    uv add numpy matplotlib jupyter
    uv sync --upgrade
    uv export -o requirements.txt --no-hashes
    uv run rsconnect write-manifest api .

    # Dev dependencies may need to be added 
    uv add --dev uv ipykernel

Deploying

Reference the Posit Connect User Guide: Shiny for Python

Using Posit Publisher

Make sure the Posit Publisher extension is installed.

Click new deployment and follow the screens to add your developer information and an API key.

Make sure to add any files that need to be included in the bundle to (1) the .toml file (2) the project files list.

For example, the .toml file might look like:

# Configuration file generated by Posit Publisher.
# Please review and modify as needed. See the documentation for more options:
# https://github.com/posit-dev/publisher/blob/main/docs/configuration.md
'$schema' = 'https://cdn.posit.co/publisher/schemas/posit-publishing-schema-v3.json'
type = 'python-shiny'
entrypoint = 'app.py'
validate = true
files = [
  '/app.py',
  '/data,csv',
  '/requirements.txt',
  '/.posit/publish/shiny-income-share-positron-FLKH.toml',
  '/.posit/publish/deployments/deployment-NVOJ.toml',
  '/data.csv'
]
title = 'shiny-income-share-positron'

[python]
version = "3.11.9"
package_file = "requirements.txt"
package_manager = "pip"

Reference: https://github.com/posit-dev/publisher/blob/main/docs/configuration.md

rsconnect-python CLI

# With uv
uv run rsconnect deploy shiny .
# Without uv
rsconnect deploy shiny .

Git-backed

Update the code, and then run:

# With uv
uv export -o requirements.txt --no-hashes
uv run rsconnect write-manifest shiny --overwrite .
# Without uv
pip freeze > requirements.txt 
rsconnect write-manifest shiny --overwrite .

Commit the new manifest.json file to the git repo along with the code.

Resources

Inspired by: https://www.andrewheiss.com/blog/2024/07/08/fun-with-positron/